Skip to content

Never stop coding. Free MIT AI gateway: one endpoint, 375 providers (150+ free), 1,400+ models Kimi, Claude, GPT, Gemini, GLM, DeepSeek, MiniMax. Works with Claude Code, Codex, Cursor, OpenCode, Cline & Copilot. Quota-aware auto-fallback, RTK+Caveman compression saves 15-95% tokens, MCP/A2A, Desktop/PWA. Built by 900+ contributors

feat(skills): preserve curated content from 18 old SKILL.md via custom-start/end markers + archive

diegosouzapw2e1862192e57
diff --git a/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-a2a/SKILL.md b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-a2a/SKILL.md
new file mode 100644
index 0000000000..2783c6261f
--- /dev/null
+++ b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-a2a/SKILL.md
@@ -0,0 +1,71 @@
+---
+name: omniroute-a2a
+description: OmniRoute exposes an A2A (Agent-to-Agent) JSON-RPC 2.0 server with 5 skills (smart-routing, quota-management, provider-discovery, cost-analysis, health-report). Use when the user wants OmniRoute to act as an agent peer in an A2A network or multi-agent pipeline.
+---
+
+# OmniRoute — A2A Protocol
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+OmniRoute publishes an Agent Card at `/.well-known/agent.json` and accepts
+JSON-RPC 2.0 calls at `/a2a`.
+
+## Discovery
+
+```bash
+curl $OMNIROUTE_URL/.well-known/agent.json
+```
+
+Returns Agent Card with skills, endpoints, auth scheme.
+
+## Available skills
+
+| Skill | Purpose |
+| -------------------- | ----------------------------------------------------------------------------------- |
+| `smart-routing` | Given a prompt, recommends best provider/model combo |
+| `quota-management` | Reports quota balance for given provider/account |
+| `provider-discovery` | Lists providers matching capability filters (vision, JSON mode, tools, max-context) |
+| `cost-analysis` | Estimates cost for a given request shape |
+| `health-report` | Returns system health (circuit states, latencies, errors) |
+
+## Call example (JSON-RPC 2.0)
+
+```bash
+curl -X POST $OMNIROUTE_URL/a2a \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "jsonrpc": "2.0",
+ "method": "tasks/send",
+ "params": {
+ "skillId": "smart-routing",
+ "input": { "prompt_length": 4000, "tools": true, "vision": false }
+ },
+ "id": 1
+ }'
+```
+
+## Response shape
+
+```json
+{
+ "jsonrpc": "2.0",
+ "result": {
+ "taskId": "...",
+ "status": "completed",
+ "output": { "recommended_combo": "...", "reasoning": "..." }
+ },
+ "id": 1
+}
+```
+
+## Errors
+
+- `-32600` → invalid request (bad JSON-RPC envelope)
+- `-32601` → method not found (check `method` field)
+- `-32602` → invalid params (check `skillId` against Agent Card)
+- `401` → missing/invalid `OMNIROUTE_KEY`
+
+## Reference
+
+Full docs: https://github.com/diegosouzapw/OmniRoute/blob/main/docs/frameworks/A2A-SERVER.md
diff --git a/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-chat/SKILL.md b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-chat/SKILL.md
new file mode 100644
index 0000000000..7f97c73cd0
--- /dev/null
+++ b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-chat/SKILL.md
@@ -0,0 +1,68 @@
+---
+name: omniroute-chat
+description: Chat / code generation via OmniRoute using OpenAI /v1/chat/completions or Anthropic /v1/messages format with SSE streaming, auto-fallback combos, RTK token saver, and 207+ providers. Use when the user wants to ask an LLM, generate code, summarize text, or run prompts through OmniRoute.
+---
+
+# OmniRoute — Chat
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+## Endpoints
+
+- `POST $OMNIROUTE_URL/v1/chat/completions` — OpenAI format
+- `POST $OMNIROUTE_URL/v1/messages` — Anthropic Messages format
+- `POST $OMNIROUTE_URL/v1/responses` — OpenAI Responses API
+
+## Discover
+
+```bash
+curl $OMNIROUTE_URL/v1/models | jq '.data[].id'
+```
+
+Combos (e.g. `auto`, `cost-optimized`, `subscription`) auto-fallback through multiple providers.
+
+## OpenAI format example
+
+```bash
+curl -X POST $OMNIROUTE_URL/v1/chat/completions \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "model": "claude-opus-4-7",
+ "messages": [{"role": "user", "content": "Refactor this function"}],
+ "stream": true
+ }'
+```
+
+## Anthropic format example
+
+```bash
+curl -X POST $OMNIROUTE_URL/v1/messages \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "anthropic-version: 2023-06-01" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "model": "claude-opus-4-7",
+ "max_tokens": 4096,
+ "messages": [{"role": "user", "content": "Hi"}]
+ }'
+```
+
+## Tool use
+
+Supports OpenAI `tools` array and Anthropic `tools` block. Tool results
+auto-compressed via RTK (47 filters: git-diff, grep, test-jest, terraform-plan,
+docker-logs, etc.) — 20-40% token savings. Disable per-request with
+`X-Omniroute-Rtk: off` header.
+
+## Reasoning / thinking
+
+Anthropic extended thinking and OpenAI Responses reasoning blocks are forwarded
+verbatim. Cached automatically via reasoning cache.
+
+## Errors
+
+- `401` → invalid API key
+- `400 invalid_model` → model not in registry; check `/v1/models`
+- `503 circuit_open` → provider circuit breaker tripped; retry later or use combo
+- `429 rate_limited` → honor `Retry-After`; consider using a combo for auto-fallback
diff --git a/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-cli-admin/SKILL.md b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-cli-admin/SKILL.md
new file mode 100644
index 0000000000..082817cf4d
--- /dev/null
+++ b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-cli-admin/SKILL.md
@@ -0,0 +1,145 @@
+---
+name: omniroute-cli-admin
+description: Manage the OmniRoute server lifecycle via CLI — start/stop/restart, non-interactive setup, diagnostics (omniroute doctor), backup/restore, autostart, and tunnel management. Use when the user wants to operate the OmniRoute server, automate provisioning, or troubleshoot the runtime.
+---
+
+# OmniRoute — CLI Admin
+
+Requires the `omniroute` CLI. See [CLI entry-point skill](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli/SKILL.md) for install + global flags.
+
+## Server lifecycle
+
+```bash
+omniroute # Start server (default port 20128)
+omniroute serve # Explicit alias
+omniroute --port 3000 # Override port
+omniroute --no-open # Don't auto-open browser
+omniroute --mcp # Start as MCP server (stdio transport)
+
+omniroute stop # Stop the running server
+omniroute restart # Restart the server
+
+omniroute dashboard # Open dashboard in browser
+omniroute open # Alias for dashboard
+omniroute status # Runtime status (uptime, requests, providers)
+```
+
+## Setup & provisioning
+
+### Interactive wizard
+
+```bash
+omniroute setup # Step-by-step interactive setup
+```
+
+### Non-interactive (CI / Docker)
+
+```bash
+omniroute setup --non-interactive \
+ --password 'admin-password' \
+ --add-provider \
+ --provider openai \
+ --api-key 'sk-...' \
+ --test-provider
+```
+
+Environment variables for non-interactive setup:
+
+| Variable | Purpose |
+| ----------------------------- | -------------------------------------------- |
+| `OMNIROUTE_SETUP_PASSWORD` | Admin password (≥8 chars) |
+| `OMNIROUTE_PROVIDER` | Provider id (e.g. `openai`, `anthropic`) |
+| `OMNIROUTE_PROVIDER_NAME` | Display name for the connection |
+| `OMNIROUTE_PROVIDER_BASE_URL` | Optional OpenAI-compatible base URL override |
+| `OMNIROUTE_API_KEY` | Provider API key |
+| `OMNIROUTE_DEFAULT_MODEL` | Optional default model |
+| `DATA_DIR` | Override OmniRoute data directory |
+
+## Diagnostics
+
+```bash
+omniroute doctor # Full health check
+omniroute doctor --json # Machine-readable JSON
+omniroute doctor --no-liveness # Skip HTTP health probe
+omniroute doctor --host 0.0.0.0 # Override liveness host
+omniroute doctor --liveness-url <url> # Full URL override
+```
+
+Checks performed: Config, Database, Storage/encryption, Port, Node runtime, Native binary (better-sqlite3), Memory, Server liveness.
+
+Exit code is non-zero if any check fails — useful in CI:
+
+```bash
+omniroute doctor --json | jq '.checks[] | select(.status=="fail")'
+```
+
+## Backup & restore
+
+```bash
+omniroute backup # Snapshot config + SQLite DB to ~/.omniroute/backups/
+omniroute restore # Restore from a previous snapshot (interactive picker)
+```
+
+## Autostart (system tray / startup)
+
+```bash
+omniroute autostart enable # Register OmniRoute as a system startup item
+omniroute autostart disable # Remove startup registration
+omniroute autostart status # Show current autostart state
+```
+
+On Linux: creates a **systemd user service** (`~/.config/systemd/user/omniroute.service`) and enables **linger** so the service can start after reboot without a graphical login; on desktop sessions it also adds an XDG autostart entry with `--tray`. On macOS: LaunchAgent plist. On Windows: registry startup entry.
+
+## Tunnels (public URL)
+
+Expose a local OmniRoute instance via a secure tunnel:
+
+```bash
+omniroute tunnel list # List active tunnels
+omniroute tunnel create cloudflare # Start a Cloudflare Tunnel (free)
+omniroute tunnel create tailscale # Start a Tailscale funnel
+omniroute tunnel create ngrok # Start an ngrok tunnel
+omniroute tunnel stop <id> # Stop a running tunnel
+```
+
+The tunnel URL is printed and can be used as `OMNIROUTE_BASE_URL` from remote machines.
+
+## Config & environment
+
+```bash
+omniroute config show # Display current effective configuration
+omniroute env show # List all OmniRoute environment variables
+omniroute env get <KEY> # Get a single env var value
+omniroute env set <KEY> <value> # Set an env var (temporary — until restart)
+```
+
+## Recovery
+
+```bash
+omniroute reset-password # Reset the admin password interactively
+omniroute reset-encrypted-columns # Dry-run: show encrypted credential columns
+omniroute reset-encrypted-columns --force # Null out encrypted credentials in SQLite
+```
+
+Use `reset-encrypted-columns --force` only if `STORAGE_ENCRYPTION_KEY` was lost and you need to re-enter all provider API keys.
+
+## Logs
+
+```bash
+omniroute logs # Stream live request logs
+omniroute logs --json # JSON log entries
+omniroute logs --search <term> # Filter by term
+omniroute logs --follow # Tail mode (keep streaming)
+```
+
+## Update
+
+```bash
+omniroute update # Check for a newer version and prompt to update
+```
+
+## Errors
+
+- `doctor` shows `STORAGE_ENCRYPTION_KEY missing` → set the key in `.env` or run `reset-encrypted-columns --force` to wipe and re-enter credentials
+- `doctor` reports native binary fail → `npm rebuild better-sqlite3` in the OmniRoute app directory
+- `tunnel create cloudflare` hangs → ensure `cloudflared` is installed: `brew install cloudflare/cloudflare/cloudflared`
diff --git a/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-cli-cloud/SKILL.md b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-cli-cloud/SKILL.md
new file mode 100644
index 0000000000..75e83ef4a6
--- /dev/null
+++ b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-cli-cloud/SKILL.md
@@ -0,0 +1,120 @@
+---
+name: omniroute-cli-cloud
+description: Control OmniRoute cloud agents (OpenAI Codex, Devin, Jules) from the CLI — create tasks, track status, approve plans, send messages, and manage sources. Use when the user wants to automate cloud coding agent workflows via the terminal.
+---
+
+# OmniRoute — CLI Cloud Agents
+
+Requires the `omniroute` CLI. See [CLI entry-point skill](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli/SKILL.md) for install + global flags.
+
+## Supported agents
+
+| Agent | ID | Auth required |
+| -------------- | ------- | -------------- |
+| OpenAI Codex | `codex` | OpenAI API key |
+| Devin | `devin` | Devin API key |
+| Jules (Google) | `jules` | Google OAuth |
+
+## Authenticate an agent
+
+```bash
+omniroute cloud codex auth # Set / refresh Codex API key
+omniroute cloud devin auth # Set / refresh Devin API key
+omniroute cloud jules auth # OAuth login for Jules (opens browser)
+```
+
+## List all agents
+
+```bash
+omniroute cloud agents # Show all configured cloud agents + status
+omniroute cloud agents --json
+```
+
+## Create a task
+
+```bash
+omniroute cloud codex task create \
+ --title "Add OAuth to the auth module" \
+ --prompt "Implement Google OAuth 2.0 login in src/auth/oauth.ts"
+
+omniroute cloud codex task create \
+ --title "Fix test failures" \
+ --prompt-file ./task-description.md # Read prompt from a file
+```
+
+Same syntax for `devin` and `jules`:
+
+```bash
+omniroute cloud devin task create --title "..." --prompt "..."
+omniroute cloud jules task create --title "..." --prompt "..."
+```
+
+## List tasks
+
+```bash
+omniroute cloud codex task list # All Codex tasks (table)
+omniroute cloud codex task list --json # JSON output
+```
+
+## Get task details
+
+```bash
+omniroute cloud codex task get <taskId> # Full task record
+omniroute cloud codex task status <taskId> # Status + progress only
+```
+
+Task status values: `running`, `completed`, `failed`, `cancelled`.
+
+## Approve a plan
+
+Devin and Jules may pause for plan approval before writing code:
+
+```bash
+omniroute cloud devin task approve <taskId> # Approve the proposed plan
+omniroute cloud jules task approve <taskId>
+```
+
+## Send a message to a running task
+
+```bash
+omniroute cloud codex task message <taskId> "Focus on the backend only, skip the UI"
+```
+
+## List task sources (files touched)
+
+```bash
+omniroute cloud codex task sources <taskId> # Files changed / created by the task
+```
+
+## Cancel a task
+
+```bash
+omniroute cloud codex task cancel <taskId>
+omniroute cloud devin task cancel <taskId>
+```
+
+## Typical workflow
+
+```bash
+# 1. Authenticate
+omniroute cloud codex auth
+
+# 2. Create a task
+TASK_ID=$(omniroute cloud codex task create \
+ --title "Refactor auth module" \
+ --prompt "Extract JWT logic to src/auth/jwt.ts" \
+ --output json | jq -r '.id')
+
+# 3. Poll status
+omniroute cloud codex task status $TASK_ID
+
+# 4. View touched files when complete
+omniroute cloud codex task sources $TASK_ID
+```
+
+## Errors
+
+- `auth required` → run `omniroute cloud <agent> auth` before any task commands
+- `task create` fails with 402 → agent billing limit reached; check your Codex/Devin/Jules account
+- `task approve` fails with 404 → task does not have a pending plan; check status first
+- `jules auth` browser flow fails → ensure Google account has Jules access; try `omniroute cloud jules auth` again
diff --git a/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-cli-eval/SKILL.md b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-cli-eval/SKILL.md
new file mode 100644
index 0000000000..244097d4b7
--- /dev/null
+++ b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-cli-eval/SKILL.md
@@ -0,0 +1,113 @@
+---
+name: omniroute-cli-eval
+description: Run and manage OmniRoute eval suites from the CLI — create suites, run benchmarks, watch live results, view scorecards, and compare model performance. Use when the user wants to benchmark models, validate quality regressions, or automate LLM evals in CI.
+---
+
+# OmniRoute — CLI Evals
+
+Requires the `omniroute` CLI. See [CLI entry-point skill](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli/SKILL.md) for install + global flags.
+
+## What are evals?
+
+Evals are automated test suites that score LLM outputs against expected answers or rubrics. OmniRoute stores suites and run results in its local database.
+
+## Eval suites
+
+```bash
+omniroute eval suites list # List all eval suites
+omniroute eval suites list --json # JSON output
+
+omniroute eval suites get <suiteId> # Full suite definition
+```
+
+### Create a suite
+
+```bash
+omniroute eval suites create \
+ --name "code-quality" \
+ --rubric "exact-match" \
+ --samples-file ./samples.jsonl # JSONL: {input, expected_output}
+```
+
+Rubric options: `exact-match`, `contains`, `llm-judge`, `regex`.
+
+`--samples-file` format (one JSON object per line):
+
+```jsonl
+{"input": "What is 2+2?", "expected_output": "4"}
+{"input": "Translate 'hello' to Spanish", "expected_output": "hola"}
+```
+
+## Run an eval
+
+```bash
+omniroute eval suites run <suiteId> \
+ --model claude-sonnet-4-6 # Run suite against a specific model
+
+omniroute eval suites run <suiteId> \
+ --model gpt-4o \
+ --watch # Live TUI progress (EvalWatch)
+```
+
+The run is asynchronous. Use `--watch` for a live terminal dashboard or poll manually:
+
+```bash
+RUN_ID=$(omniroute eval suites run <suiteId> --model claude-sonnet-4-6 --output json | jq -r '.id')
+omniroute eval get $RUN_ID
+```
+
+## Manage runs
+
+```bash
+omniroute eval list # List all eval runs
+omniroute eval list --json
+
+omniroute eval get <runId> # Run details (status, model, score)
+omniroute eval results <runId> # Per-sample results
+omniroute eval scorecard <runId> # Full scorecard with pass/fail per sample
+omniroute eval cancel <runId> # Cancel a running eval
+```
+
+## Scorecard output
+
+```bash
+omniroute eval scorecard <runId> --output json
+```
+
+Response fields per sample:
+
+```json
+{
+ "id": "sample-1",
+ "score": 0.95,
+ "passed": true,
+ "input": "What is 2+2?",
+ "output": "4",
+ "expected": "4"
+}
+```
+
+## Comparing models
+
+Run the same suite against multiple models and compare:
+
+```bash
+for MODEL in claude-sonnet-4-6 gpt-4o gemini-2.0-flash; do
+ omniroute eval suites run $SUITE_ID --model $MODEL --output json | jq '{model: .model, score: .score}'
+done
+```
+
+## CI integration
+
+```bash
+# Run and fail CI if score drops below threshold
+SCORE=$(omniroute eval suites run $SUITE_ID --model claude-sonnet-4-6 --output json | jq -r '.score')
+python3 -c "import sys; score=float('$SCORE'); sys.exit(0 if score >= 0.90 else 1)"
+```
+
+## Errors
+
+- `suites create` fails with `invalid rubric` → use one of: `exact-match`, `contains`, `llm-judge`, `regex`
+- `suites run` returns `model not found` → verify model ID with `omniroute models --search <name>`
+- `eval get` shows `status: failed` → check `omniroute logs --search eval` for error details
+- `scorecard` returns empty results → the run may still be `running`; poll `omniroute eval get <runId>` until `status` is `completed`
diff --git a/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-cli-providers/SKILL.md b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-cli-providers/SKILL.md
new file mode 100644
index 0000000000..1d8e53a5b9
--- /dev/null
+++ b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-cli-providers/SKILL.md
@@ -0,0 +1,145 @@
+---
+name: omniroute-cli-providers
+description: Manage OmniRoute provider connections, API keys, and routing combos via CLI — add/list/test/remove providers, rotate keys, run OAuth flows, list models, and create/switch combos. Use when the user wants to configure providers, manage credentials, or set up routing from the terminal.
+---
+
+# OmniRoute — CLI Providers & Keys
+
+Requires the `omniroute` CLI. See [CLI entry-point skill](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli/SKILL.md) for install + global flags.
+
+## Provider catalog (available providers)
+
+```bash
+omniroute providers available # Full OmniRoute provider catalog
+omniroute providers available --search openai # Filter by id, name, alias
+omniroute providers available --category api-key # Filter by category
+omniroute providers available --json # Machine-readable JSON
+```
+
+Categories: `api-key`, `oauth`, `free`, `local`, `combo`.
+
+## Configured provider connections
+
+```bash
+omniroute providers list # Connections in your DB
+omniroute providers list --json
+```
+
+## Testing connections
+
+```bash
+omniroute providers test <id|name> # Test one configured connection
+omniroute providers test-all # Test every active connection (TUI progress)
+omniroute providers validate # Local-only structural validation (no HTTP)
+```
+
+`test-all` opens an interactive TUI that shows live pass/fail per connection. Use `--json` to get a machine-readable result:
+
+```bash
+omniroute providers test-all --json
+```
+
+## API key management (OmniRoute keys)
+
+These manage the OmniRoute API keys issued under **API Manager** — not provider credentials.
+
+```bash
+omniroute keys list # List all OmniRoute API keys
+omniroute keys add <provider> [apiKey] # Add an API key for a provider
+omniroute keys remove <provider> # Remove an API key
+omniroute keys regenerate <id> # Regenerate (rotate) a key
+omniroute keys revoke <id> # Revoke a key (disables it)
+omniroute keys reveal <id> # Show the full key value
+omniroute keys usage <id> # Show usage stats for a key
+
+omniroute keys rotate <id> # Rotate + revoke old key atomically
+omniroute keys expiration list # List key expiration times
+```
+
+### Key policies
+
+```bash
+omniroute keys policy show <id> # Show rate-limit / permission policy
+omniroute keys policy set <id> \
+ --rate-limit 100 \
+ --rate-window minute \
+ --permissions chat,models # Set policy on a key
+```
+
+## Models
+
+```bash
+omniroute models # List all models (all providers)
+omniroute models openai # Filter by provider
+omniroute models --search gpt # Search by name
+omniroute models --json # JSON output
+```
+
+## OAuth providers
+
+```bash
+omniroute oauth list # List OAuth-configured providers
+omniroute oauth login <provider> # Start browser-based OAuth flow
+omniroute oauth logout <provider> # Revoke OAuth token
+omniroute oauth status <provider> # Show token state + expiry
+omniroute oauth refresh <provider> # Force token refresh
+```
+
+For OAuth providers (Gemini, Windsurf, Antigravity, etc.) the `login` command opens the OmniRoute dashboard OAuth flow in your browser.
+
+## Provider nodes (multi-account routing)
+
+Provider nodes let you attach multiple API keys / accounts to one logical provider for round-robin or failover.
+
+```bash
+omniroute nodes list <provider> # List nodes for a provider
+omniroute nodes add <provider> --api-key <key> # Add a node
+omniroute nodes remove <provider> <nodeId> # Remove a node
+omniroute nodes test <provider> <nodeId> # Test one node
+```
+
+## Routing combos (CLI)
+
+Create and manage routing combos from the terminal:
+
+```bash
+omniroute combo list # List all combos
+omniroute combo create <name> \
+ --strategy priority \
+ --targets anthropic/claude-opus-4-7,openai/gpt-4o # Create combo
+omniroute combo switch <name> # Activate a combo as default
+omniroute combo delete <name> # Delete a combo
+omniroute combo suggest --task "code review" # Ask OmniRoute to recommend a combo
+```
+
+For the full REST API for combos see [omniroute-routing skill](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-routing/SKILL.md).
+
+## Quota & usage
+
+```bash
+omniroute quota # Provider quota usage + reset times
+omniroute usage # Request + token usage summary
+omniroute cost # Cost breakdown (by provider/model)
+```
+
+## Compression (CLI)
+
+```bash
+omniroute compression status # Current compression mode + savings stats
+omniroute compression set --mode rtk # Enable RTK compression
+omniroute compression set --mode stacked # Enable stacked (RTK + Caveman)
+omniroute compression set --mode off # Disable compression
+omniroute compression preview --mode rtk --text "..." # Preview savings for sample text
+```
+
+## Health
+
+```bash
+omniroute health # Detailed health: circuit breakers, cache, memory
+```
+
+## Errors
+
+- `providers test <id>` fails with 401 → API key stored for that provider is wrong; use `keys remove` + `keys add` to reset it
+- `oauth login` opens but doesn't complete → the OAuth token endpoint may be firewalled; use API key auth instead
+- `combo create` fails with `strategy unknown` → use one of: `priority`, `weighted`, `round-robin`, `fill-first`, `least-used`, `cost-optimized`, `auto`, `random`, `strict-random`, `p2c`, `reset-aware`, `lkgp`, `context-optimized`, `context-relay`
diff --git a/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-cli/SKILL.md b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-cli/SKILL.md
new file mode 100644
index 0000000000..5229916395
--- /dev/null
+++ b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-cli/SKILL.md
@@ -0,0 +1,104 @@
+---
+name: omniroute-cli
+description: Entry point for the OmniRoute CLI (omniroute binary) — install, global flags, output formats, environment variables, and index of CLI capability skills. Use when the user wants to control OmniRoute from the terminal, automate workflows, or integrate with CI/CD.
+---
+
+# OmniRoute — CLI (Entry Point)
+
+The `omniroute` binary ships with the OmniRoute server. It is both the server launcher and a full management CLI with 250+ commands across 39 groups.
+
+## Install
+
+```bash
+npm install -g omniroute # npm registry
+# or: use the binary bundled with the desktop app
+```
+
+Requires Node.js ≥20.20.2, ≥22.22.2, or ≥24.
+
+Verify:
+
+```bash
+omniroute --version # prints installed version
+omniroute --help # full command tree
+```
+
+## Connection
+
+Every CLI command that talks to the server reads two values:
+
+| Source | Variable / Flag |
+| -------- | ------------------------------------ |
+| Base URL | `OMNIROUTE_BASE_URL` or `--base-url` |
+| API key | `OMNIROUTE_API_KEY` or `--api-key` |
+
+Default base URL: `http://localhost:20128`
+
+```bash
+export OMNIROUTE_BASE_URL="http://localhost:20128"
+export OMNIROUTE_API_KEY="sk-..." # from Dashboard → API Manager
+```
+
+For a remote server:
+
+```bash
+export OMNIROUTE_BASE_URL="https://your-server.com"
+```
+
+## Global flags
+
+| Flag | Description |
+| ------------------- | -------------------------------------------------------- |
+| `--base-url <url>` | Override server URL for this invocation |
+| `--api-key <key>` | Override API key for this invocation |
+| `--output <format>` | Output format: `table` (default), `json`, `jsonl`, `csv` |
+| `--json` | Shorthand for `--output json` |
+| `--non-interactive` | Disable prompts — for CI / shell scripts |
+| `--no-open` | Don't auto-open the browser on start |
+| `--port <n>` | Override default port 20128 |
+| `--help`, `-h` | Show help for the current command |
+| `--version`, `-v` | Print the installed version |
+
+## Output formats
+
+All listing commands support `--output`:
+
+```bash
+omniroute combo list # human-readable table
+omniroute combo list --output json # JSON array
+omniroute combo list --output jsonl # one JSON object per line
+omniroute combo list --output csv # CSV with header row
+```
+
+## Quick start: one-shot server + provider setup
+
+```bash
+# 1. Start server
+omniroute
+
+# 2. (First run) interactive setup wizard
+omniroute setup
+
+# 3. Verify everything is healthy
+omniroute doctor
+```
+
+## CLI capability skills
+
+| Capability | Skill |
+| ------------------------------------ | ----------------------------------------------------------------------------------------------------- |
+| Server admin + backup | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-admin/SKILL.md |
+| Provider & key management | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-providers/SKILL.md |
+| Cloud agents (Codex / Devin / Jules) | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-cloud/SKILL.md |
+| Evals & benchmarking | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-eval/SKILL.md |
+
+## API skills (REST)
+
+For direct HTTP usage instead of the CLI, see the [OmniRoute entry-point skill](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md).
+
+## Errors
+
+- `Connection refused` → server not running; run `omniroute` or `omniroute serve`
+- `401 Unauthorized` → wrong or missing API key
+- `command not found: omniroute` → not in PATH; check `npm root -g` or re-install
+- `doctor` reports SQLite incompatible → `npm rebuild better-sqlite3` in the app directory
diff --git a/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-compression/SKILL.md b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-compression/SKILL.md
new file mode 100644
index 0000000000..fd670ead69
--- /dev/null
+++ b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-compression/SKILL.md
@@ -0,0 +1,131 @@
+---
+name: omniroute-compression
+description: Configure OmniRoute token compression to save 60–90% of context tokens. Covers RTK (command/tool output), Caveman (prose), stacked mode (both), and the MCP accessibility-tree filter (browser snapshots). Use when the user wants to reduce costs, fit long sessions into context windows, or speed up AI responses.
+---
+
+# OmniRoute — Compression
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+## Overview
+
+OmniRoute compresses token payloads before forwarding to providers. No code changes required — set it once, it applies to all requests transparently.
+
+| Engine | Best for | Typical savings |
+| ------------------------- | ------------------------------------ | --------------- |
+| RTK | Terminal / build / test / git output | 60–90% |
+| Caveman | Human prose, chat history | 46% input |
+| Stacked (`rtk → caveman`) | Mixed coding sessions | 78–95% |
+| MCP accessibility filter | Browser/accessibility tool results | 60–80% |
+
+## Get current settings
+
+```bash
+curl $OMNIROUTE_URL/api/settings/compression \
+ -H "Authorization: Bearer $OMNIROUTE_KEY"
+```
+
+## Enable RTK (best for coding agents)
+
+```bash
+curl -X PUT $OMNIROUTE_URL/api/settings/compression \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{ "mode": "rtk", "enabled": true }'
+```
+
+## Enable stacked mode (maximum savings)
+
+```bash
+curl -X PUT $OMNIROUTE_URL/api/settings/compression \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "mode": "stacked",
+ "enabled": true,
+ "stackedPipeline": ["rtk", "caveman"]
+ }'
+```
+
+## Enable Caveman (prose / chat)
+
+```bash
+curl -X PUT $OMNIROUTE_URL/api/settings/compression \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{ "mode": "standard", "enabled": true }'
+```
+
+Caveman intensities: `lite` (safe), `standard` (balanced), `aggressive` (long sessions), `ultra` (context recovery).
+
+## Preview compression before enabling
+
+```bash
+curl -X POST $OMNIROUTE_URL/api/compression/preview \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "mode": "rtk",
+ "text": "$ npm test\n> jest\n\nPASS src/a.test.ts (2.1s)\nPASS src/b.test.ts (1.8s)\n..."
+ }'
+```
+
+Response includes `compressed`, `original_length`, `compressed_length`, `savings_pct`.
+
+## MCP accessibility-tree filter (browser agent use)
+
+When OmniRoute is used with browser/Playwright MCP tools, it automatically compresses verbose accessibility-tree tool results. Enabled by default; configure thresholds:
+
+```bash
+curl -X PUT $OMNIROUTE_URL/api/settings/compression \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "mcpAccessibility": {
+ "enabled": true,
+ "collapseThreshold": 30,
+ "maxTextChars": 50000
+ }
+ }'
+```
+
+`collapseThreshold`: collapse sibling lines when ≥ N repeats (default 30).
+`maxTextChars`: hard truncate after N chars with navigation hint (default 50000).
+
+## Language packs (Caveman)
+
+Caveman supports language-aware rules for pt-BR, es, de, fr, ja:
+
+```bash
+curl -X PUT $OMNIROUTE_URL/api/settings/compression \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "mode": "standard",
+ "cavemanConfig": {
+ "language": "pt-BR",
+ "autoDetectLanguage": true
+ }
+ }'
+```
+
+## Via MCP
+
+```
+omniroute_compression_status → current settings + savings analytics
+omniroute_compression_configure → update mode/threshold/language
+omniroute_set_compression_engine → switch engine at runtime
+```
+
+## Disable compression
+
+```bash
+curl -X PUT $OMNIROUTE_URL/api/settings/compression \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -d '{ "enabled": false }'
+```
+
+## Errors
+
+- `400 invalid mode` → use `off`, `lite`, `standard`, `aggressive`, `ultra`, `rtk`, or `stacked`
+- `400 invalid stackedPipeline` → array must contain valid engine ids (`rtk`, `caveman`)
diff --git a/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-embeddings/SKILL.md b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-embeddings/SKILL.md
new file mode 100644
index 0000000000..f02998092e
--- /dev/null
+++ b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-embeddings/SKILL.md
@@ -0,0 +1,45 @@
+---
+name: omniroute-embeddings
+description: Embeddings via OmniRoute using OpenAI /v1/embeddings format with auto-fallback across text-embedding-3-large, Voyage, Cohere, Gemini embeddings, Jina. Use when the user needs vector embeddings for RAG, similarity search, or clustering.
+---
+
+# OmniRoute — Embeddings
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+## Endpoint
+
+- `POST $OMNIROUTE_URL/v1/embeddings`
+
+## Discover
+
+```bash
+curl $OMNIROUTE_URL/v1/models/embedding | jq '.data[]'
+```
+
+Each entry: `{ id, owned_by, dimensions, max_input_tokens }`.
+
+## Example
+
+```bash
+curl -X POST $OMNIROUTE_URL/v1/embeddings \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "model": "text-embedding-3-large",
+ "input": ["first text", "second text"],
+ "encoding_format": "float"
+ }'
+```
+
+Response: `{ data:[{ embedding:[...], index }], usage:{ prompt_tokens, total_tokens } }`
+
+## Batch input
+
+`input` accepts a string or array of strings (up to provider batch limit, typically 2048 items).
+
+## Errors
+
+- `400 input_too_long` → input exceeds `max_input_tokens` for this model
+- `400 invalid_encoding_format` → use `float` or `base64`
+- `503` → provider unavailable; try another model in `/v1/models/embedding`
diff --git a/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-image/SKILL.md b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-image/SKILL.md
new file mode 100644
index 0000000000..9ccd40a170
--- /dev/null
+++ b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-image/SKILL.md
@@ -0,0 +1,45 @@
+---
+name: omniroute-image
+description: Image generation via OmniRoute using OpenAI /v1/images/generations format with auto-fallback across DALL-E, Stable Diffusion, Flux, Imagen providers. Use when the user wants to generate, edit, or vary images.
+---
+
+# OmniRoute — Image Generation
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+## Endpoints
+
+- `POST $OMNIROUTE_URL/v1/images/generations` — Text-to-image
+- `POST $OMNIROUTE_URL/v1/images/edits` — Image edit (mask)
+- `POST $OMNIROUTE_URL/v1/images/variations` — Variations
+
+## Discover
+
+```bash
+curl $OMNIROUTE_URL/v1/models/image | jq '.data[]'
+```
+
+Returns `{ id, owned_by, sizes:[...], capabilities:[...] }` per model.
+
+## Generate example
+
+```bash
+curl -X POST $OMNIROUTE_URL/v1/images/generations \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "model": "dall-e-3",
+ "prompt": "a red bicycle on a wet street, photoreal",
+ "n": 1,
+ "size": "1024x1024",
+ "response_format": "b64_json"
+ }'
+```
+
+Response: `{ created, data: [{ url? or b64_json, revised_prompt }] }`
+
+## Errors
+
+- `400 invalid_size` → not supported by this model; check `/v1/models/image`
+- `400 content_policy_violation` → blocked by provider safety
+- `503` → provider unavailable; try another model in `/v1/models/image`
diff --git a/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-mcp/SKILL.md b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-mcp/SKILL.md
new file mode 100644
index 0000000000..594896d1b1
--- /dev/null
+++ b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-mcp/SKILL.md
@@ -0,0 +1,71 @@
+---
+name: omniroute-mcp
+description: OmniRoute exposes a built-in MCP (Model Context Protocol) server with 37 tools (chat, embeddings, memory CRUD, skills, providers, routing, audit) over SSE/stdio/HTTP transports. Use when the user wants to add OmniRoute as an MCP server in Claude Desktop, Cursor, Cline, or any MCP-compatible client.
+---
+
+# OmniRoute — MCP Server
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+## Transports
+
+- **stdio** — local IPC, for Claude Desktop / VS Code extensions
+- **SSE** — `GET $OMNIROUTE_URL/api/mcp/sse`
+- **Streamable HTTP** — `POST $OMNIROUTE_URL/api/mcp/stream`
+
+## Claude Desktop config
+
+Add to `~/Library/Application Support/Claude/claude_desktop_config.json`:
+
+```json
+{
+ "mcpServers": {
+ "omniroute": {
+ "command": "npx",
+ "args": ["-y", "omniroute", "--mcp"],
+ "env": { "OMNIROUTE_KEY": "sk-..." }
+ }
+ }
+}
+```
+
+## Cursor / VS Code config
+
+```json
+{
+ "mcp": {
+ "servers": {
+ "omniroute": {
+ "url": "http://localhost:20128/api/mcp/sse",
+ "headers": { "Authorization": "Bearer sk-..." }
+ }
+ }
+ }
+}
+```
+
+## Available tools (37 total)
+
+| Scope | Tools |
+| --------- | -------------------------------------------------------------------------------------------------- |
+| health | `omniroute_get_health` |
+| combos | `omniroute_list_combos`, `omniroute_get_combo_metrics`, `omniroute_switch_combo` |
+| routing | `omniroute_simulate_route`, `omniroute_best_combo_for_task`, `omniroute_explain_route` |
+| providers | `omniroute_get_provider_metrics`, `omniroute_check_quota`, `omniroute_route_request` |
+| budget | `omniroute_set_budget_guard`, `omniroute_set_routing_strategy`, `omniroute_set_resilience_profile` |
+| testing | `omniroute_test_combo` |
+| memory | `memory_add`, `memory_search`, `memory_delete` |
+| skills | `skill_invoke`, `skill_list`, `skill_describe`, `skill_register` |
+| cache | `omniroute_cache_stats`, `omniroute_cache_flush` |
+| admin | `omniroute_db_health_check`, `omniroute_sync_pricing`, `omniroute_get_session_snapshot` |
+
+Full list: `GET $OMNIROUTE_URL/api/mcp/tools`
+
+## Scopes
+
+Tools are grouped into 13 scopes (chat-only, memory-readonly, full-admin, etc.).
+Pass scope name as `--scope` arg or via `X-Omniroute-Scope` header.
+
+## Reference
+
+Full docs: https://github.com/diegosouzapw/OmniRoute/blob/main/docs/frameworks/MCP-SERVER.md
diff --git a/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-monitoring/SKILL.md b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-monitoring/SKILL.md
new file mode 100644
index 0000000000..f2f0b23c08
--- /dev/null
+++ b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-monitoring/SKILL.md
@@ -0,0 +1,114 @@
+---
+name: omniroute-monitoring
+description: Monitor OmniRoute system health, provider circuit breakers, per-provider latency (p50/p95/p99), quota usage, and set budget guards. Use when the user wants to check if the system is healthy, debug slow providers, manage spend limits, or set up oncall-style monitoring.
+---
+
+# OmniRoute — Monitoring & Health
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+## System health
+
+```bash
+curl $OMNIROUTE_URL/api/health \
+ -H "Authorization: Bearer $OMNIROUTE_KEY"
+```
+
+Returns: uptime, memory, active connections, circuit breaker states, rate limit status, cache stats.
+
+Unauthenticated quick check:
+
+```bash
+curl $OMNIROUTE_URL/api/health
+# → {"ok":true}
+```
+
+## Provider circuit breakers
+
+Circuit breakers prevent traffic from hitting failing providers.
+
+States: `CLOSED` (normal), `OPEN` (blocked), `HALF_OPEN` (probe mode — auto-recovers).
+
+```bash
+curl $OMNIROUTE_URL/api/monitoring/health \
+ -H "Authorization: Bearer $OMNIROUTE_KEY"
+```
+
+Response includes `circuitBreakers` array with per-provider state and `resetAt` timestamp.
+
+## Per-provider metrics (p50/p95/p99)
+
+```bash
+curl $OMNIROUTE_URL/api/providers/metrics \
+ -H "Authorization: Bearer $OMNIROUTE_KEY"
+```
+
+Response shape per provider:
+
+```json
+{
+ "provider": "anthropic",
+ "requests": 1247,
+ "successRate": 0.994,
+ "latency": { "p50": 820, "p95": 2100, "p99": 3800 },
+ "circuitState": "CLOSED",
+ "tokensUsed": 2847000
+}
+```
+
+## Via MCP (if OmniRoute is your MCP server)
+
+```
+omniroute_get_health → full system health snapshot
+omniroute_get_provider_metrics → p50/p95/p99 + circuit state per provider
+omniroute_get_session_snapshot → cost, tokens, errors for current session
+omniroute_check_quota → quota balance + percent remaining + reset time
+omniroute_db_health_check → diagnose + auto-repair database drift
+```
+
+## Quota check
+
+```bash
+curl $OMNIROUTE_URL/api/quota \
+ -H "Authorization: Bearer $OMNIROUTE_KEY"
+```
+
+Returns used/total tokens and requests per provider/account, with `resetAt` timestamps.
+
+## Budget guard (spend limit)
+
+Set a session spending limit that degrades or blocks requests when hit:
+
+```bash
+curl -X POST $OMNIROUTE_URL/api/budget/guard \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "limitUsd": 5.00,
+ "action": "degrade",
+ "degradeTo": "openai/gpt-4o-mini"
+ }'
+```
+
+`action` options:
+
+- `degrade` — switch to a cheaper model when limit is hit
+- `block` — return 429 when limit is hit
+- `alert` — continue but add `X-Budget-Warning` header
+
+## MCP audit log
+
+OmniRoute logs every MCP tool call to `mcp_audit` table. Query via API:
+
+```bash
+curl "$OMNIROUTE_URL/api/mcp/status" \
+ -H "Authorization: Bearer $OMNIROUTE_KEY"
+```
+
+Returns: server status, heartbeat, recent audit activity summary.
+
+## Errors
+
+- `503` on health endpoint → OmniRoute is starting up; retry in 5s
+- Circuit breaker `OPEN` → provider is temporarily blocked; check `resetAt` to know when it auto-recovers
+- `429 budget_exceeded` → budget guard limit reached; raise limit or wait for reset
diff --git a/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-routing/SKILL.md b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-routing/SKILL.md
new file mode 100644
index 0000000000..d0a90de766
--- /dev/null
+++ b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-routing/SKILL.md
@@ -0,0 +1,129 @@
+---
+name: omniroute-routing
+description: Create and configure OmniRoute routing combos, choose from 14 strategies (priority, weighted, auto, round-robin, cost-optimized, etc.), activate Auto-combo 9-factor scoring, and set up fallback chains. Use when the user wants to configure multi-provider routing, load balancing, or cost-optimized model selection.
+---
+
+# OmniRoute — Routing & Combos
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+## What is a combo?
+
+A combo is a named group of providers/models with a routing strategy. All requests through a combo are automatically distributed, failed-over, and load-balanced — the caller uses a single model ID like `my-combo`.
+
+## List existing combos
+
+```bash
+curl $OMNIROUTE_URL/api/combos \
+ -H "Authorization: Bearer $OMNIROUTE_KEY"
+```
+
+Response includes `id`, `name`, `strategy`, `enabled`, and per-target stats.
+
+## Create a combo
+
+```bash
+curl -X POST $OMNIROUTE_URL/api/combos \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "name": "my-combo",
+ "strategy": "priority",
+ "targets": [
+ { "provider": "anthropic", "model": "claude-opus-4-7", "weight": 1 },
+ { "provider": "openai", "model": "gpt-4o", "weight": 1 }
+ ]
+ }'
+```
+
+## 14 routing strategies
+
+| Strategy | Description |
+| ------------------- | -------------------------------------------------- |
+| `priority` | Always use target[0]; fall back on error |
+| `weighted` | Distribute by weight percentage |
+| `round-robin` | Rotate targets in order |
+| `fill-first` | Fill quota of target[0] before spilling |
+| `least-used` | Route to target with fewest active requests |
+| `cost-optimized` | Pick cheapest target for the token estimate |
+| `auto` | 9-factor scoring: cost + latency + quota + circuit |
+| `random` | Uniform random selection |
+| `strict-random` | Random without repeating until all used |
+| `p2c` | Power-of-2-choices: sample 2, pick better |
+| `reset-aware` | Prefer targets near quota reset time |
+| `lkgp` | Last-known-good-provider sticky routing |
+| `context-optimized` | Pick best model for context length |
+| `context-relay` | Chain models for very long contexts |
+
+## Auto-combo (recommended for production)
+
+Auto-combo scores each candidate on 9 factors every request:
+
+```bash
+curl -X POST $OMNIROUTE_URL/api/combos \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "name": "prod-auto",
+ "strategy": "auto",
+ "targets": [
+ { "provider": "anthropic", "model": "claude-sonnet-4-6" },
+ { "provider": "openai", "model": "gpt-4o-mini" },
+ { "provider": "google", "model": "gemini-2.0-flash" }
+ ]
+ }'
+```
+
+Then call it with:
+
+```bash
+curl -X POST $OMNIROUTE_URL/v1/chat/completions \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{ "model": "prod-auto", "messages": [{ "role": "user", "content": "Hello" }] }'
+```
+
+## Activate / deactivate a combo
+
+```bash
+# Activate
+curl -X PUT $OMNIROUTE_URL/api/combos/{id}/toggle \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -d '{ "enabled": true }'
+```
+
+## Get combo metrics
+
+```bash
+curl $OMNIROUTE_URL/api/combos/{id}/metrics \
+ -H "Authorization: Bearer $OMNIROUTE_KEY"
+```
+
+Returns p50/p95/p99 latency, success rate, cost, and per-target breakdown.
+
+## Simulate routing (dry run)
+
+```bash
+curl -X POST $OMNIROUTE_URL/api/routing/simulate \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{ "comboId": "{id}", "messages": [{ "role": "user", "content": "test" }] }'
+```
+
+Returns which provider would be selected and why — no actual API call is made.
+
+## Via MCP (if OmniRoute is your MCP server)
+
+```
+omniroute_list_combos → list all combos
+omniroute_switch_combo → enable/disable a combo
+omniroute_set_routing_strategy → change strategy at runtime
+omniroute_simulate_route → dry-run routing decision
+omniroute_best_combo_for_task → get recommendation by task type
+```
+
+## Errors
+
+- `404 combo not found` → check `id` from `/api/combos`
+- `400 invalid strategy` → use one of the 14 strategies above
+- `409 name conflict` → combo name already exists
diff --git a/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-stt/SKILL.md b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-stt/SKILL.md
new file mode 100644
index 0000000000..4008b69ffb
--- /dev/null
+++ b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-stt/SKILL.md
@@ -0,0 +1,42 @@
+---
+name: omniroute-stt
+description: Speech-to-text via OmniRoute using OpenAI /v1/audio/transcriptions format with auto-fallback across Whisper, AssemblyAI, Deepgram, Azure STT. Use when the user wants transcription of audio files or real-time speech recognition.
+---
+
+# OmniRoute — Speech-to-Text
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+## Endpoints
+
+- `POST $OMNIROUTE_URL/v1/audio/transcriptions` — multipart upload, returns text
+- `POST $OMNIROUTE_URL/v1/audio/translations` — transcribe + translate to English
+
+## Discover
+
+```bash
+curl $OMNIROUTE_URL/v1/models/stt | jq '.data[]'
+```
+
+## Example
+
+```bash
+curl -X POST $OMNIROUTE_URL/v1/audio/transcriptions \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -F "model=whisper-1" \
+ -F "response_format=verbose_json"
+```
+
+Response: `{ text, language, duration, segments?:[{ start, end, text }] }`
+
+## Supported formats
+
+Audio: `mp3`, `mp4`, `mpeg`, `mpga`, `m4a`, `wav`, `webm`.
+Response formats: `json`, `text`, `srt`, `verbose_json`, `vtt`.
+
+## Errors
+
+- `400 invalid_file_format` → unsupported audio format
+- `400 file_too_large` → exceeds provider limit (usually 25MB)
+- `503` → provider unavailable; try another model in `/v1/models/stt`
diff --git a/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-tts/SKILL.md b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-tts/SKILL.md
new file mode 100644
index 0000000000..f236615ca5
--- /dev/null
+++ b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-tts/SKILL.md
@@ -0,0 +1,45 @@
+---
+name: omniroute-tts
+description: Text-to-speech via OmniRoute using OpenAI /v1/audio/speech format with auto-fallback across OpenAI TTS, ElevenLabs, Azure Neural, Google Cloud TTS. Use when the user wants spoken audio output from text.
+---
+
+# OmniRoute — Text-to-Speech
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+## Endpoint
+
+- `POST $OMNIROUTE_URL/v1/audio/speech` — returns binary audio (mp3/opus/wav/flac)
+
+## Discover
+
+```bash
+curl $OMNIROUTE_URL/v1/models/tts | jq '.data[]'
+```
+
+Each entry includes `voices:[...]` for the available voice names per provider.
+
+## Example
+
+```bash
+curl -X POST $OMNIROUTE_URL/v1/audio/speech \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "model": "tts-1",
+ "input": "Hello from OmniRoute.",
+ "voice": "alloy",
+ "response_format": "mp3"
+ }' --output speech.mp3
+```
+
+## Voices
+
+Voice names vary by provider. Check `/v1/models/tts` — each entry has `voices:[...]`.
+Common OpenAI voices: `alloy`, `echo`, `fable`, `onyx`, `nova`, `shimmer`.
+
+## Errors
+
+- `400 invalid_voice` → voice not supported by this model
+- `400 input_too_long` → input exceeds model character limit
+- `503` → provider unavailable; try another model in `/v1/models/tts`
diff --git a/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-web-fetch/SKILL.md b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-web-fetch/SKILL.md
new file mode 100644
index 0000000000..9c009e1954
--- /dev/null
+++ b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-web-fetch/SKILL.md
@@ -0,0 +1,47 @@
+---
+name: omniroute-web-fetch
+description: Fetch a URL and convert to clean markdown via OmniRoute proxying Jina Reader, Firecrawl, raw HTML strip. Use when the user wants to ingest a webpage as markdown for context in an LLM conversation.
+---
+
+# OmniRoute — Web Fetch
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+## Endpoint
+
+- `POST $OMNIROUTE_URL/v1/web/fetch`
+
+## Discover
+
+```bash
+curl $OMNIROUTE_URL/v1/models/web | jq '.data[] | select(.kind == "webFetch")'
+```
+
+## Example
+
+```bash
+curl -X POST $OMNIROUTE_URL/v1/web/fetch \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "model": "jina/reader",
+ "url": "https://anthropic.com",
+ "format": "markdown"
+ }'
+```
+
+Response: `{ url, title, markdown, links?:[...], images?:[...] }`
+
+## Parameters
+
+| Field | Type | Description |
+| -------- | ------ | ----------------------------------------------------------------------- |
+| `model` | string | Provider from `/v1/models/web` (e.g. `jina/reader`, `firecrawl/scrape`) |
+| `url` | string | URL to fetch |
+| `format` | string | `markdown` (default), `html`, `text` |
+
+## Errors
+
+- `400 invalid_url` → URL must be http/https
+- `403 blocked` → provider blocked by target site; try a different model
+- `503` → provider unavailable; try another model in `/v1/models/web`
diff --git a/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-web-search/SKILL.md b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-web-search/SKILL.md
new file mode 100644
index 0000000000..ca169b2305
--- /dev/null
+++ b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute-web-search/SKILL.md
@@ -0,0 +1,49 @@
+---
+name: omniroute-web-search
+description: Web search via OmniRoute proxying Tavily, Brave Search, SerpAPI, Exa with auto-fallback. Use when the user wants live web search results, current news, or facts that may be beyond the LLM training cutoff.
+---
+
+# OmniRoute — Web Search
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+## Endpoint
+
+- `POST $OMNIROUTE_URL/v1/web/search` — unified search format
+
+## Discover
+
+```bash
+curl $OMNIROUTE_URL/v1/models/web | jq '.data[] | select(.kind == "webSearch")'
+```
+
+## Example
+
+```bash
+curl -X POST $OMNIROUTE_URL/v1/web/search \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "model": "tavily/search",
+ "query": "OmniRoute github latest release",
+ "max_results": 5,
+ "include_answer": true
+ }'
+```
+
+Response: `{ answer?, results:[{ url, title, content, score }] }`
+
+## Parameters
+
+| Field | Type | Description |
+| ---------------- | ------- | ------------------------------------ |
+| `model` | string | Provider model from `/v1/models/web` |
+| `query` | string | Search query |
+| `max_results` | number | Max results (default: 5) |
+| `include_answer` | boolean | Include AI-synthesized answer |
+| `search_depth` | string | `basic` or `advanced` (Tavily) |
+
+## Errors
+
+- `400 query_too_long` → shorten the search query
+- `503` → provider unavailable; try another model in `/v1/models/web`
diff --git a/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute/SKILL.md b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute/SKILL.md
new file mode 100644
index 0000000000..6291d47a53
--- /dev/null
+++ b/_tasks/features-v3.8.6/refactorpages/_orchestration/15-pruned-archive/omniroute/SKILL.md
@@ -0,0 +1,76 @@
+---
+name: omniroute
+description: Entry point for OmniRoute — local/remote AI gateway with OpenAI-compatible REST for chat, image, TTS, STT, embeddings, web search, web fetch, MCP, A2A. Use when the user mentions OmniRoute, OMNIROUTE_URL, or wants AI without writing provider boilerplate. This skill covers setup + indexes capability skills; fetch the relevant capability SKILL.md from the URLs below when needed.
+---
+
+# OmniRoute
+
+Local/remote AI gateway exposing OpenAI-compatible REST. One key, 207+ providers,
+auto-fallback, RTK token saver, MCP server, A2A agents.
+
+## Setup
+
+```bash
+export OMNIROUTE_URL="http://localhost:20128" # or VPS / tunnel URL
+export OMNIROUTE_KEY="sk-..." # from Dashboard → API Keys
+```
+
+All requests: `${OMNIROUTE_URL}/v1/...` with `Authorization: Bearer ${OMNIROUTE_KEY}`.
+
+Verify: `curl $OMNIROUTE_URL/api/health` → `{"ok":true}`
+
+## Discover models
+
+```bash
+curl $OMNIROUTE_URL/v1/models # chat/LLM (default)
+curl $OMNIROUTE_URL/v1/models/image # image-gen
+curl $OMNIROUTE_URL/v1/models/tts # text-to-speech
+curl $OMNIROUTE_URL/v1/models/embedding # embeddings
+curl $OMNIROUTE_URL/v1/models/web # web search + fetch
+curl $OMNIROUTE_URL/v1/models/stt # speech-to-text
+```
+
+Use `data[].id` as `model` field in requests. Combos appear with `owned_by:"combo"`.
+
+## Capability skills
+
+| Capability | Raw URL |
+| --------------------- | --------------------------------------------------------------------------------------------------- |
+| Chat / code-gen | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-chat/SKILL.md |
+| Image generation | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-image/SKILL.md |
+| Text-to-speech | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-tts/SKILL.md |
+| Speech-to-text | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-stt/SKILL.md |
+| Embeddings | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-embeddings/SKILL.md |
+| Web search | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-web-search/SKILL.md |
+| Web fetch | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-web-fetch/SKILL.md |
+| MCP server (37 tools) | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-mcp/SKILL.md |
+| A2A protocol | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-a2a/SKILL.md |
+| Routing & combos | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-routing/SKILL.md |
+| Token compression | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-compression/SKILL.md |
+| Monitoring & health | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-monitoring/SKILL.md |
+
+## CLI skills (omniroute binary)
+
+| Capability | Raw URL |
+| ---------------------- | ----------------------------------------------------------------------------------------------------- |
+| CLI entry point | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli/SKILL.md |
+| CLI admin & lifecycle | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-admin/SKILL.md |
+| CLI providers & keys | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-providers/SKILL.md |
+| CLI cloud agents | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-cloud/SKILL.md |
+| CLI evals & benchmarks | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-eval/SKILL.md |
+
+## Errors
+
+- `401` → set/refresh `OMNIROUTE_KEY` (Dashboard → API Keys)
+- `400 Invalid model format` → check `model` exists in `/v1/models/<kind>`
+- `503 Provider circuit open` → upstream provider down; retry after `Retry-After` seconds
+- `429` → rate limited; honor `Retry-After`
+
+## Differentiators vs OpenAI direct
+
+- **Auto-fallback** combos (14 strategies): never stop coding even if a provider rate-limits
+- **RTK token saver**: tool_result compressed via 47 specialized filters (git-diff, test-jest, terraform-plan, docker-logs…) — 20-40% token reduction
+- **Caveman mode**: optional terse system prompt injection (LITE/FULL/ULTRA) — 15-25% completion reduction
+- **MCP + A2A** servers built-in (this is the only AI router that exposes both protocols)
+- **Memory** with FTS5 + Qdrant for persistent agent context
+- **Guardrails** for PII masking, prompt injection detection, vision policies
diff --git a/skills/cli-eval/SKILL.md b/skills/cli-eval/SKILL.md
new file mode 100644
index 0000000000..769deff4be
--- /dev/null
+++ b/skills/cli-eval/SKILL.md
@@ -0,0 +1,117 @@
+---
+name: cli-eval
+description: Run and manage OmniRoute eval suites from the CLI — create suites, run benchmarks, watch live results, view scorecards, and compare model performance.
+---
+
+<!-- skill:custom-start -->
+<!-- Migrated from skills/omniroute-cli-eval/SKILL.md (preserved curated content) -->
+
+# OmniRoute — CLI Evals
+
+Requires the `omniroute` CLI. See [CLI entry-point skill](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli/SKILL.md) for install + global flags.
+
+## What are evals?
+
+Evals are automated test suites that score LLM outputs against expected answers or rubrics. OmniRoute stores suites and run results in its local database.
+
+## Eval suites
+
+```bash
+omniroute eval suites list # List all eval suites
+omniroute eval suites list --json # JSON output
+
+omniroute eval suites get <suiteId> # Full suite definition
+```
+
+### Create a suite
+
+```bash
+omniroute eval suites create \
+ --name "code-quality" \
+ --rubric "exact-match" \
+ --samples-file ./samples.jsonl # JSONL: {input, expected_output}
+```
+
+Rubric options: `exact-match`, `contains`, `llm-judge`, `regex`.
+
+`--samples-file` format (one JSON object per line):
+
+```jsonl
+{"input": "What is 2+2?", "expected_output": "4"}
+{"input": "Translate 'hello' to Spanish", "expected_output": "hola"}
+```
+
+## Run an eval
+
+```bash
+omniroute eval suites run <suiteId> \
+ --model claude-sonnet-4-6 # Run suite against a specific model
+
+omniroute eval suites run <suiteId> \
+ --model gpt-4o \
+ --watch # Live TUI progress (EvalWatch)
+```
+
+The run is asynchronous. Use `--watch` for a live terminal dashboard or poll manually:
+
+```bash
+RUN_ID=$(omniroute eval suites run <suiteId> --model claude-sonnet-4-6 --output json | jq -r '.id')
+omniroute eval get $RUN_ID
+```
+
+## Manage runs
+
+```bash
+omniroute eval list # List all eval runs
+omniroute eval list --json
+
+omniroute eval get <runId> # Run details (status, model, score)
+omniroute eval results <runId> # Per-sample results
+omniroute eval scorecard <runId> # Full scorecard with pass/fail per sample
+omniroute eval cancel <runId> # Cancel a running eval
+```
+
+## Scorecard output
+
+```bash
+omniroute eval scorecard <runId> --output json
+```
+
+Response fields per sample:
+
+```json
+{
+ "id": "sample-1",
+ "score": 0.95,
+ "passed": true,
+ "input": "What is 2+2?",
+ "output": "4",
+ "expected": "4"
+}
+```
+
+## Comparing models
+
+Run the same suite against multiple models and compare:
+
+```bash
+for MODEL in claude-sonnet-4-6 gpt-4o gemini-2.0-flash; do
+ omniroute eval suites run $SUITE_ID --model $MODEL --output json | jq '{model: .model, score: .score}'
+done
+```
+
+## CI integration
+
+```bash
+# Run and fail CI if score drops below threshold
+SCORE=$(omniroute eval suites run $SUITE_ID --model claude-sonnet-4-6 --output json | jq -r '.score')
+python3 -c "import sys; score=float('$SCORE'); sys.exit(0 if score >= 0.90 else 1)"
+```
+
+## Errors
+
+- `suites create` fails with `invalid rubric` → use one of: `exact-match`, `contains`, `llm-judge`, `regex`
+- `suites run` returns `model not found` → verify model ID with `omniroute models --search <name>`
+- `eval get` shows `status: failed` → check `omniroute logs --search eval` for error details
+- `scorecard` returns empty results → the run may still be `running`; poll `omniroute eval get <runId>` until `status` is `completed`
+<!-- skill:custom-end -->
diff --git a/skills/cli-providers/SKILL.md b/skills/cli-providers/SKILL.md
new file mode 100644
index 0000000000..c89747b23e
--- /dev/null
+++ b/skills/cli-providers/SKILL.md
@@ -0,0 +1,149 @@
+---
+name: cli-providers
+description: Manage OmniRoute provider connections, API keys, and routing combos via CLI — add/list/test/remove providers, rotate keys, run OAuth flows, list models, and create/switch combos.
+---
+
+<!-- skill:custom-start -->
+<!-- Migrated from skills/omniroute-cli-providers/SKILL.md (preserved curated content) -->
+
+# OmniRoute — CLI Providers & Keys
+
+Requires the `omniroute` CLI. See [CLI entry-point skill](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli/SKILL.md) for install + global flags.
+
+## Provider catalog (available providers)
+
+```bash
+omniroute providers available # Full OmniRoute provider catalog
+omniroute providers available --search openai # Filter by id, name, alias
+omniroute providers available --category api-key # Filter by category
+omniroute providers available --json # Machine-readable JSON
+```
+
+Categories: `api-key`, `oauth`, `free`, `local`, `combo`.
+
+## Configured provider connections
+
+```bash
+omniroute providers list # Connections in your DB
+omniroute providers list --json
+```
+
+## Testing connections
+
+```bash
+omniroute providers test <id|name> # Test one configured connection
+omniroute providers test-all # Test every active connection (TUI progress)
+omniroute providers validate # Local-only structural validation (no HTTP)
+```
+
+`test-all` opens an interactive TUI that shows live pass/fail per connection. Use `--json` to get a machine-readable result:
+
+```bash
+omniroute providers test-all --json
+```
+
+## API key management (OmniRoute keys)
+
+These manage the OmniRoute API keys issued under **API Manager** — not provider credentials.
+
+```bash
+omniroute keys list # List all OmniRoute API keys
+omniroute keys add <provider> [apiKey] # Add an API key for a provider
+omniroute keys remove <provider> # Remove an API key
+omniroute keys regenerate <id> # Regenerate (rotate) a key
+omniroute keys revoke <id> # Revoke a key (disables it)
+omniroute keys reveal <id> # Show the full key value
+omniroute keys usage <id> # Show usage stats for a key
+
+omniroute keys rotate <id> # Rotate + revoke old key atomically
+omniroute keys expiration list # List key expiration times
+```
+
+### Key policies
+
+```bash
+omniroute keys policy show <id> # Show rate-limit / permission policy
+omniroute keys policy set <id> \
+ --rate-limit 100 \
+ --rate-window minute \
+ --permissions chat,models # Set policy on a key
+```
+
+## Models
+
+```bash
+omniroute models # List all models (all providers)
+omniroute models openai # Filter by provider
+omniroute models --search gpt # Search by name
+omniroute models --json # JSON output
+```
+
+## OAuth providers
+
+```bash
+omniroute oauth list # List OAuth-configured providers
+omniroute oauth login <provider> # Start browser-based OAuth flow
+omniroute oauth logout <provider> # Revoke OAuth token
+omniroute oauth status <provider> # Show token state + expiry
+omniroute oauth refresh <provider> # Force token refresh
+```
+
+For OAuth providers (Gemini, Windsurf, Antigravity, etc.) the `login` command opens the OmniRoute dashboard OAuth flow in your browser.
+
+## Provider nodes (multi-account routing)
+
+Provider nodes let you attach multiple API keys / accounts to one logical provider for round-robin or failover.
+
+```bash
+omniroute nodes list <provider> # List nodes for a provider
+omniroute nodes add <provider> --api-key <key> # Add a node
+omniroute nodes remove <provider> <nodeId> # Remove a node
+omniroute nodes test <provider> <nodeId> # Test one node
+```
+
+## Routing combos (CLI)
+
+Create and manage routing combos from the terminal:
+
+```bash
+omniroute combo list # List all combos
+omniroute combo create <name> \
+ --strategy priority \
+ --targets anthropic/claude-opus-4-7,openai/gpt-4o # Create combo
+omniroute combo switch <name> # Activate a combo as default
+omniroute combo delete <name> # Delete a combo
+omniroute combo suggest --task "code review" # Ask OmniRoute to recommend a combo
+```
+
+For the full REST API for combos see [omniroute-routing skill](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-routing/SKILL.md).
+
+## Quota & usage
+
+```bash
+omniroute quota # Provider quota usage + reset times
+omniroute usage # Request + token usage summary
+omniroute cost # Cost breakdown (by provider/model)
+```
+
+## Compression (CLI)
+
+```bash
+omniroute compression status # Current compression mode + savings stats
+omniroute compression set --mode rtk # Enable RTK compression
+omniroute compression set --mode stacked # Enable stacked (RTK + Caveman)
+omniroute compression set --mode off # Disable compression
+omniroute compression preview --mode rtk --text "..." # Preview savings for sample text
+```
+
+## Health
+
+```bash
+omniroute health # Detailed health: circuit breakers, cache, memory
+```
+
+## Errors
+
+- `providers test <id>` fails with 401 → API key stored for that provider is wrong; use `keys remove` + `keys add` to reset it
+- `oauth login` opens but doesn't complete → the OAuth token endpoint may be firewalled; use API key auth instead
+- `combo create` fails with `strategy unknown` → use one of: `priority`, `weighted`, `round-robin`, `fill-first`, `least-used`, `cost-optimized`, `auto`, `random`, `strict-random`, `p2c`, `reset-aware`, `lkgp`, `context-optimized`, `context-relay`
+<!-- skill:custom-end -->
diff --git a/skills/cli-serve/SKILL.md b/skills/cli-serve/SKILL.md
new file mode 100644
index 0000000000..ec638366d0
--- /dev/null
+++ b/skills/cli-serve/SKILL.md
@@ -0,0 +1,245 @@
+---
+name: cli-serve
+description: OmniRoute CLI entry point and server admin — install, global flags, output formats, server lifecycle (start/stop/restart), setup wizard, diagnostics, backup/restore, autostart, tunnels, and CI/Docker provisioning.
+---
+
+<!-- skill:custom-start -->
+<!-- Aggregated from: omniroute-cli (setup + index), omniroute-cli-admin (lifecycle/setup/backup/tunnel) -->
+
+## Setup
+
+The `omniroute` binary ships with the OmniRoute server. It is both the server launcher and a full management CLI with 250+ commands across 39 groups.
+
+### Install
+
+```bash
+npm install -g omniroute # npm registry
+# or: use the binary bundled with the desktop app
+```
+
+Requires Node.js ≥20.20.2, ≥22.22.2, or ≥24.
+
+Verify:
+
+```bash
+omniroute --version # prints installed version
+omniroute --help # full command tree
+```
+
+### Connection
+
+Every CLI command that talks to the server reads two values:
+
+| Source | Variable / Flag |
+| -------- | ------------------------------------ |
+| Base URL | `OMNIROUTE_BASE_URL` or `--base-url` |
+| API key | `OMNIROUTE_API_KEY` or `--api-key` |
+
+Default base URL: `http://localhost:20128`
+
+```bash
+export OMNIROUTE_BASE_URL="http://localhost:20128"
+export OMNIROUTE_API_KEY="sk-..." # from Dashboard → API Manager
+```
+
+For a remote server:
+
+```bash
+export OMNIROUTE_BASE_URL="https://your-server.com"
+```
+
+### Global flags
+
+| Flag | Description |
+| ------------------- | -------------------------------------------------------- |
+| `--base-url <url>` | Override server URL for this invocation |
+| `--api-key <key>` | Override API key for this invocation |
+| `--output <format>` | Output format: `table` (default), `json`, `jsonl`, `csv` |
+| `--json` | Shorthand for `--output json` |
+| `--non-interactive` | Disable prompts — for CI / shell scripts |
+| `--no-open` | Don't auto-open the browser on start |
+| `--port <n>` | Override default port 20128 |
+| `--help`, `-h` | Show help for the current command |
+| `--version`, `-v` | Print the installed version |
+
+### Output formats
+
+All listing commands support `--output`:
+
+```bash
+omniroute combo list # human-readable table
+omniroute combo list --output json # JSON array
+omniroute combo list --output jsonl # one JSON object per line
+omniroute combo list --output csv # CSV with header row
+```
+
+### Quick start: one-shot server + provider setup
+
+```bash
+# 1. Start server
+omniroute
+
+# 2. (First run) interactive setup wizard
+omniroute setup
+
+# 3. Verify everything is healthy
+omniroute doctor
+```
+
+### CLI capability skills
+
+| Capability | Skill |
+| ------------------------------------ | ----------------------------------------------------------------------------------------------------- |
+| Server admin + backup | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-admin/SKILL.md |
+| Provider & key management | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-providers/SKILL.md |
+| Cloud agents (Codex / Devin / Jules) | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-cloud/SKILL.md |
+| Evals & benchmarking | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-eval/SKILL.md |
+
+### Errors
+
+- `Connection refused` → server not running; run `omniroute` or `omniroute serve`
+- `401 Unauthorized` → wrong or missing API key
+- `command not found: omniroute` → not in PATH; check `npm root -g` or re-install
+- `doctor` reports SQLite incompatible → `npm rebuild better-sqlite3` in the app directory
+
+## Admin lifecycle
+
+Requires the `omniroute` CLI. See [CLI entry-point skill](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli/SKILL.md) for install + global flags.
+
+### Server lifecycle
+
+```bash
+omniroute # Start server (default port 20128)
+omniroute serve # Explicit alias
+omniroute --port 3000 # Override port
+omniroute --no-open # Don't auto-open browser
+omniroute --mcp # Start as MCP server (stdio transport)
+
+omniroute stop # Stop the running server
+omniroute restart # Restart the server
+
+omniroute dashboard # Open dashboard in browser
+omniroute open # Alias for dashboard
+omniroute status # Runtime status (uptime, requests, providers)
+```
+
+### Setup & provisioning
+
+#### Interactive wizard
+
+```bash
+omniroute setup # Step-by-step interactive setup
+```
+
+#### Non-interactive (CI / Docker)
+
+```bash
+omniroute setup --non-interactive \
+ --password 'admin-password' \
+ --add-provider \
+ --provider openai \
+ --api-key 'sk-...' \
+ --test-provider
+```
+
+Environment variables for non-interactive setup:
+
+| Variable | Purpose |
+| ----------------------------- | -------------------------------------------- |
+| `OMNIROUTE_SETUP_PASSWORD` | Admin password (≥8 chars) |
+| `OMNIROUTE_PROVIDER` | Provider id (e.g. `openai`, `anthropic`) |
+| `OMNIROUTE_PROVIDER_NAME` | Display name for the connection |
+| `OMNIROUTE_PROVIDER_BASE_URL` | Optional OpenAI-compatible base URL override |
+| `OMNIROUTE_API_KEY` | Provider API key |
+| `OMNIROUTE_DEFAULT_MODEL` | Optional default model |
+| `DATA_DIR` | Override OmniRoute data directory |
+
+### Diagnostics
+
+```bash
+omniroute doctor # Full health check
+omniroute doctor --json # Machine-readable JSON
+omniroute doctor --no-liveness # Skip HTTP health probe
+omniroute doctor --host 0.0.0.0 # Override liveness host
+omniroute doctor --liveness-url <url> # Full URL override
+```
+
+Checks performed: Config, Database, Storage/encryption, Port, Node runtime, Native binary (better-sqlite3), Memory, Server liveness.
+
+Exit code is non-zero if any check fails — useful in CI:
+
+```bash
+omniroute doctor --json | jq '.checks[] | select(.status=="fail")'
+```
+
+### Backup & restore
+
+```bash
+omniroute backup # Snapshot config + SQLite DB to ~/.omniroute/backups/
+omniroute restore # Restore from a previous snapshot (interactive picker)
+```
+
+### Autostart (system tray / startup)
+
+```bash
+omniroute autostart enable # Register OmniRoute as a system startup item
+omniroute autostart disable # Remove startup registration
+omniroute autostart status # Show current autostart state
+```
+
+On Linux: creates a **systemd user service** (`~/.config/systemd/user/omniroute.service`) and enables **linger** so the service can start after reboot without a graphical login; on desktop sessions it also adds an XDG autostart entry with `--tray`. On macOS: LaunchAgent plist. On Windows: registry startup entry.
+
+### Tunnels (public URL)
+
+Expose a local OmniRoute instance via a secure tunnel:
+
+```bash
+omniroute tunnel list # List active tunnels
+omniroute tunnel create cloudflare # Start a Cloudflare Tunnel (free)
+omniroute tunnel create tailscale # Start a Tailscale funnel
+omniroute tunnel create ngrok # Start an ngrok tunnel
+omniroute tunnel stop <id> # Stop a running tunnel
+```
+
+The tunnel URL is printed and can be used as `OMNIROUTE_BASE_URL` from remote machines.
+
+### Config & environment
+
+```bash
+omniroute config show # Display current effective configuration
+omniroute env show # List all OmniRoute environment variables
+omniroute env get <KEY> # Get a single env var value
+omniroute env set <KEY> <value> # Set an env var (temporary — until restart)
+```
+
+### Recovery
+
+```bash
+omniroute reset-password # Reset the admin password interactively
+omniroute reset-encrypted-columns # Dry-run: show encrypted credential columns
+omniroute reset-encrypted-columns --force # Null out encrypted credentials in SQLite
+```
+
+Use `reset-encrypted-columns --force` only if `STORAGE_ENCRYPTION_KEY` was lost and you need to re-enter all provider API keys.
+
+### Logs
+
+```bash
+omniroute logs # Stream live request logs
+omniroute logs --json # JSON log entries
+omniroute logs --search <term> # Filter by term
+omniroute logs --follow # Tail mode (keep streaming)
+```
+
+### Update
+
+```bash
+omniroute update # Check for a newer version and prompt to update
+```
+
+### Errors
+
+- `doctor` shows `STORAGE_ENCRYPTION_KEY missing` → set the key in `.env` or run `reset-encrypted-columns --force` to wipe and re-enter credentials
+- `doctor` reports native binary fail → `npm rebuild better-sqlite3` in the OmniRoute app directory
+- `tunnel create cloudflare` hangs → ensure `cloudflared` is installed: `brew install cloudflare/cloudflare/cloudflared`
+<!-- skill:custom-end -->
diff --git a/skills/omni-agents-a2a/SKILL.md b/skills/omni-agents-a2a/SKILL.md
new file mode 100644
index 0000000000..135edb3022
--- /dev/null
+++ b/skills/omni-agents-a2a/SKILL.md
@@ -0,0 +1,75 @@
+---
+name: omni-agents-a2a
+description: OmniRoute exposes an A2A (Agent-to-Agent) JSON-RPC 2.0 server with 5 skills (smart-routing, quota-management, provider-discovery, cost-analysis, health-report). Use when the user wants OmniRoute to act as an agent peer in an A2A network or multi-agent pipeline.
+---
+
+<!-- skill:custom-start -->
+<!-- Migrated from skills/omniroute-a2a/SKILL.md (preserved curated content) -->
+
+# OmniRoute — A2A Protocol
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+OmniRoute publishes an Agent Card at `/.well-known/agent.json` and accepts
+JSON-RPC 2.0 calls at `/a2a`.
+
+## Discovery
+
+```bash
+curl $OMNIROUTE_URL/.well-known/agent.json
+```
+
+Returns Agent Card with skills, endpoints, auth scheme.
+
+## Available skills
+
+| Skill | Purpose |
+| -------------------- | ----------------------------------------------------------------------------------- |
+| `smart-routing` | Given a prompt, recommends best provider/model combo |
+| `quota-management` | Reports quota balance for given provider/account |
+| `provider-discovery` | Lists providers matching capability filters (vision, JSON mode, tools, max-context) |
+| `cost-analysis` | Estimates cost for a given request shape |
+| `health-report` | Returns system health (circuit states, latencies, errors) |
+
+## Call example (JSON-RPC 2.0)
+
+```bash
+curl -X POST $OMNIROUTE_URL/a2a \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "jsonrpc": "2.0",
+ "method": "tasks/send",
+ "params": {
+ "skillId": "smart-routing",
+ "input": { "prompt_length": 4000, "tools": true, "vision": false }
+ },
+ "id": 1
+ }'
+```
+
+## Response shape
+
+```json
+{
+ "jsonrpc": "2.0",
+ "result": {
+ "taskId": "...",
+ "status": "completed",
+ "output": { "recommended_combo": "...", "reasoning": "..." }
+ },
+ "id": 1
+}
+```
+
+## Errors
+
+- `-32600` → invalid request (bad JSON-RPC envelope)
+- `-32601` → method not found (check `method` field)
+- `-32602` → invalid params (check `skillId` against Agent Card)
+- `401` → missing/invalid `OMNIROUTE_KEY`
+
+## Reference
+
+Full docs: https://github.com/diegosouzapw/OmniRoute/blob/main/docs/frameworks/A2A-SERVER.md
+<!-- skill:custom-end -->
diff --git a/skills/omni-auth/SKILL.md b/skills/omni-auth/SKILL.md
new file mode 100644
index 0000000000..03666cd04c
--- /dev/null
+++ b/skills/omni-auth/SKILL.md
@@ -0,0 +1,80 @@
+---
+name: omni-auth
+description: Entry point for OmniRoute — local/remote AI gateway with OpenAI-compatible REST for chat, image, TTS, STT, embeddings, web search, web fetch, MCP, A2A. Setup, index of capability skills, and differentiators.
+---
+
+<!-- skill:custom-start -->
+<!-- Migrated from skills/omniroute/SKILL.md (preserved curated content) -->
+
+# OmniRoute
+
+Local/remote AI gateway exposing OpenAI-compatible REST. One key, 207+ providers,
+auto-fallback, RTK token saver, MCP server, A2A agents.
+
+## Setup
+
+```bash
+export OMNIROUTE_URL="http://localhost:20128" # or VPS / tunnel URL
+export OMNIROUTE_KEY="sk-..." # from Dashboard → API Keys
+```
+
+All requests: `${OMNIROUTE_URL}/v1/...` with `Authorization: Bearer ${OMNIROUTE_KEY}`.
+
+Verify: `curl $OMNIROUTE_URL/api/health` → `{"ok":true}`
+
+## Discover models
+
+```bash
+curl $OMNIROUTE_URL/v1/models # chat/LLM (default)
+curl $OMNIROUTE_URL/v1/models/image # image-gen
+curl $OMNIROUTE_URL/v1/models/tts # text-to-speech
+curl $OMNIROUTE_URL/v1/models/embedding # embeddings
+curl $OMNIROUTE_URL/v1/models/web # web search + fetch
+curl $OMNIROUTE_URL/v1/models/stt # speech-to-text
+```
+
+Use `data[].id` as `model` field in requests. Combos appear with `owned_by:"combo"`.
+
+## Capability skills
+
+| Capability | Raw URL |
+| --------------------- | --------------------------------------------------------------------------------------------------- |
+| Chat / code-gen | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-chat/SKILL.md |
+| Image generation | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-image/SKILL.md |
+| Text-to-speech | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-tts/SKILL.md |
+| Speech-to-text | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-stt/SKILL.md |
+| Embeddings | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-embeddings/SKILL.md |
+| Web search | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-web-search/SKILL.md |
+| Web fetch | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-web-fetch/SKILL.md |
+| MCP server (37 tools) | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-mcp/SKILL.md |
+| A2A protocol | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-a2a/SKILL.md |
+| Routing & combos | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-routing/SKILL.md |
+| Token compression | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-compression/SKILL.md |
+| Monitoring & health | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-monitoring/SKILL.md |
+
+## CLI skills (omniroute binary)
+
+| Capability | Raw URL |
+| ---------------------- | ----------------------------------------------------------------------------------------------------- |
+| CLI entry point | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli/SKILL.md |
+| CLI admin & lifecycle | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-admin/SKILL.md |
+| CLI providers & keys | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-providers/SKILL.md |
+| CLI cloud agents | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-cloud/SKILL.md |
+| CLI evals & benchmarks | https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-eval/SKILL.md |
+
+## Errors
+
+- `401` → set/refresh `OMNIROUTE_KEY` (Dashboard → API Keys)
+- `400 Invalid model format` → check `model` exists in `/v1/models/<kind>`
+- `503 Provider circuit open` → upstream provider down; retry after `Retry-After` seconds
+- `429` → rate limited; honor `Retry-After`
+
+## Differentiators vs OpenAI direct
+
+- **Auto-fallback** combos (14 strategies): never stop coding even if a provider rate-limits
+- **RTK token saver**: tool_result compressed via 47 specialized filters (git-diff, test-jest, terraform-plan, docker-logs…) — 20-40% token reduction
+- **Caveman mode**: optional terse system prompt injection (LITE/FULL/ULTRA) — 15-25% completion reduction
+- **MCP + A2A** servers built-in (this is the only AI router that exposes both protocols)
+- **Memory** with FTS5 + Qdrant for persistent agent context
+- **Guardrails** for PII masking, prompt injection detection, vision policies
+<!-- skill:custom-end -->
diff --git a/skills/omni-combos-routing/SKILL.md b/skills/omni-combos-routing/SKILL.md
new file mode 100644
index 0000000000..a9033a672c
--- /dev/null
+++ b/skills/omni-combos-routing/SKILL.md
@@ -0,0 +1,133 @@
+---
+name: omni-combos-routing
+description: Create and configure OmniRoute routing combos, choose from 14 strategies (priority, weighted, auto, round-robin, cost-optimized, etc.), activate Auto-combo 9-factor scoring, and set up fallback chains.
+---
+
+<!-- skill:custom-start -->
+<!-- Migrated from skills/omniroute-routing/SKILL.md (preserved curated content) -->
+
+# OmniRoute — Routing & Combos
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+## What is a combo?
+
+A combo is a named group of providers/models with a routing strategy. All requests through a combo are automatically distributed, failed-over, and load-balanced — the caller uses a single model ID like `my-combo`.
+
+## List existing combos
+
+```bash
+curl $OMNIROUTE_URL/api/combos \
+ -H "Authorization: Bearer $OMNIROUTE_KEY"
+```
+
+Response includes `id`, `name`, `strategy`, `enabled`, and per-target stats.
+
+## Create a combo
+
+```bash
+curl -X POST $OMNIROUTE_URL/api/combos \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "name": "my-combo",
+ "strategy": "priority",
+ "targets": [
+ { "provider": "anthropic", "model": "claude-opus-4-7", "weight": 1 },
+ { "provider": "openai", "model": "gpt-4o", "weight": 1 }
+ ]
+ }'
+```
+
+## 14 routing strategies
+
+| Strategy | Description |
+| ------------------- | -------------------------------------------------- |
+| `priority` | Always use target[0]; fall back on error |
+| `weighted` | Distribute by weight percentage |
+| `round-robin` | Rotate targets in order |
+| `fill-first` | Fill quota of target[0] before spilling |
+| `least-used` | Route to target with fewest active requests |
+| `cost-optimized` | Pick cheapest target for the token estimate |
+| `auto` | 9-factor scoring: cost + latency + quota + circuit |
+| `random` | Uniform random selection |
+| `strict-random` | Random without repeating until all used |
+| `p2c` | Power-of-2-choices: sample 2, pick better |
+| `reset-aware` | Prefer targets near quota reset time |
+| `lkgp` | Last-known-good-provider sticky routing |
+| `context-optimized` | Pick best model for context length |
+| `context-relay` | Chain models for very long contexts |
+
+## Auto-combo (recommended for production)
+
+Auto-combo scores each candidate on 9 factors every request:
+
+```bash
+curl -X POST $OMNIROUTE_URL/api/combos \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "name": "prod-auto",
+ "strategy": "auto",
+ "targets": [
+ { "provider": "anthropic", "model": "claude-sonnet-4-6" },
+ { "provider": "openai", "model": "gpt-4o-mini" },
+ { "provider": "google", "model": "gemini-2.0-flash" }
+ ]
+ }'
+```
+
+Then call it with:
+
+```bash
+curl -X POST $OMNIROUTE_URL/v1/chat/completions \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{ "model": "prod-auto", "messages": [{ "role": "user", "content": "Hello" }] }'
+```
+
+## Activate / deactivate a combo
+
+```bash
+# Activate
+curl -X PUT $OMNIROUTE_URL/api/combos/{id}/toggle \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -d '{ "enabled": true }'
+```
+
+## Get combo metrics
+
+```bash
+curl $OMNIROUTE_URL/api/combos/{id}/metrics \
+ -H "Authorization: Bearer $OMNIROUTE_KEY"
+```
+
+Returns p50/p95/p99 latency, success rate, cost, and per-target breakdown.
+
+## Simulate routing (dry run)
+
+```bash
+curl -X POST $OMNIROUTE_URL/api/routing/simulate \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{ "comboId": "{id}", "messages": [{ "role": "user", "content": "test" }] }'
+```
+
+Returns which provider would be selected and why — no actual API call is made.
+
+## Via MCP (if OmniRoute is your MCP server)
+
+```
+omniroute_list_combos → list all combos
+omniroute_switch_combo → enable/disable a combo
+omniroute_set_routing_strategy → change strategy at runtime
+omniroute_simulate_route → dry-run routing decision
+omniroute_best_combo_for_task → get recommendation by task type
+```
+
+## Errors
+
+- `404 combo not found` → check `id` from `/api/combos`
+- `400 invalid strategy` → use one of the 14 strategies above
+- `409 name conflict` → combo name already exists
+<!-- skill:custom-end -->
diff --git a/skills/omni-compression/SKILL.md b/skills/omni-compression/SKILL.md
new file mode 100644
index 0000000000..ce80c4a920
--- /dev/null
+++ b/skills/omni-compression/SKILL.md
@@ -0,0 +1,135 @@
+---
+name: omni-compression
+description: Configure OmniRoute token compression to save 60-90% of context tokens. Covers RTK (command/tool output), Caveman (prose), stacked mode (both), and the MCP accessibility-tree filter (browser snapshots).
+---
+
+<!-- skill:custom-start -->
+<!-- Migrated from skills/omniroute-compression/SKILL.md (preserved curated content) -->
+
+# OmniRoute — Compression
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+## Overview
+
+OmniRoute compresses token payloads before forwarding to providers. No code changes required — set it once, it applies to all requests transparently.
+
+| Engine | Best for | Typical savings |
+| ------------------------- | ------------------------------------ | --------------- |
+| RTK | Terminal / build / test / git output | 60–90% |
+| Caveman | Human prose, chat history | 46% input |
+| Stacked (`rtk → caveman`) | Mixed coding sessions | 78–95% |
+| MCP accessibility filter | Browser/accessibility tool results | 60–80% |
+
+## Get current settings
+
+```bash
+curl $OMNIROUTE_URL/api/settings/compression \
+ -H "Authorization: Bearer $OMNIROUTE_KEY"
+```
+
+## Enable RTK (best for coding agents)
+
+```bash
+curl -X PUT $OMNIROUTE_URL/api/settings/compression \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{ "mode": "rtk", "enabled": true }'
+```
+
+## Enable stacked mode (maximum savings)
+
+```bash
+curl -X PUT $OMNIROUTE_URL/api/settings/compression \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "mode": "stacked",
+ "enabled": true,
+ "stackedPipeline": ["rtk", "caveman"]
+ }'
+```
+
+## Enable Caveman (prose / chat)
+
+```bash
+curl -X PUT $OMNIROUTE_URL/api/settings/compression \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{ "mode": "standard", "enabled": true }'
+```
+
+Caveman intensities: `lite` (safe), `standard` (balanced), `aggressive` (long sessions), `ultra` (context recovery).
+
+## Preview compression before enabling
+
+```bash
+curl -X POST $OMNIROUTE_URL/api/compression/preview \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "mode": "rtk",
+ "text": "$ npm test\n> jest\n\nPASS src/a.test.ts (2.1s)\nPASS src/b.test.ts (1.8s)\n..."
+ }'
+```
+
+Response includes `compressed`, `original_length`, `compressed_length`, `savings_pct`.
+
+## MCP accessibility-tree filter (browser agent use)
+
+When OmniRoute is used with browser/Playwright MCP tools, it automatically compresses verbose accessibility-tree tool results. Enabled by default; configure thresholds:
+
+```bash
+curl -X PUT $OMNIROUTE_URL/api/settings/compression \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "mcpAccessibility": {
+ "enabled": true,
+ "collapseThreshold": 30,
+ "maxTextChars": 50000
+ }
+ }'
+```
+
+`collapseThreshold`: collapse sibling lines when ≥ N repeats (default 30).
+`maxTextChars`: hard truncate after N chars with navigation hint (default 50000).
+
+## Language packs (Caveman)
+
+Caveman supports language-aware rules for pt-BR, es, de, fr, ja:
+
+```bash
+curl -X PUT $OMNIROUTE_URL/api/settings/compression \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "mode": "standard",
+ "cavemanConfig": {
+ "language": "pt-BR",
+ "autoDetectLanguage": true
+ }
+ }'
+```
+
+## Via MCP
+
+```
+omniroute_compression_status → current settings + savings analytics
+omniroute_compression_configure → update mode/threshold/language
+omniroute_set_compression_engine → switch engine at runtime
+```
+
+## Disable compression
+
+```bash
+curl -X PUT $OMNIROUTE_URL/api/settings/compression \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -d '{ "enabled": false }'
+```
+
+## Errors
+
+- `400 invalid mode` → use `off`, `lite`, `standard`, `aggressive`, `ultra`, `rtk`, or `stacked`
+- `400 invalid stackedPipeline` → array must contain valid engine ids (`rtk`, `caveman`)
+<!-- skill:custom-end -->
diff --git a/skills/omni-inference/SKILL.md b/skills/omni-inference/SKILL.md
new file mode 100644
index 0000000000..2d2d03aaff
--- /dev/null
+++ b/skills/omni-inference/SKILL.md
@@ -0,0 +1,321 @@
+---
+name: omni-inference
+description: Inference capabilities via OmniRoute — chat completions, image generation, text-to-speech, speech-to-text, embeddings, web search, and web fetch. OpenAI-compatible endpoints with auto-fallback across 207+ providers.
+---
+
+<!-- skill:custom-start -->
+<!-- Aggregated from: omniroute-chat, omniroute-image, omniroute-tts, omniroute-stt, omniroute-embeddings, omniroute-web-search, omniroute-web-fetch -->
+
+## Chat completions
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+### Endpoints
+
+- `POST $OMNIROUTE_URL/v1/chat/completions` — OpenAI format
+- `POST $OMNIROUTE_URL/v1/messages` — Anthropic Messages format
+- `POST $OMNIROUTE_URL/v1/responses` — OpenAI Responses API
+
+### Discover
+
+```bash
+curl $OMNIROUTE_URL/v1/models | jq '.data[].id'
+```
+
+Combos (e.g. `auto`, `cost-optimized`, `subscription`) auto-fallback through multiple providers.
+
+### OpenAI format example
+
+```bash
+curl -X POST $OMNIROUTE_URL/v1/chat/completions \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "model": "claude-opus-4-7",
+ "messages": [{"role": "user", "content": "Refactor this function"}],
+ "stream": true
+ }'
+```
+
+### Anthropic format example
+
+```bash
+curl -X POST $OMNIROUTE_URL/v1/messages \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "anthropic-version: 2023-06-01" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "model": "claude-opus-4-7",
+ "max_tokens": 4096,
+ "messages": [{"role": "user", "content": "Hi"}]
+ }'
+```
+
+### Tool use
+
+Supports OpenAI `tools` array and Anthropic `tools` block. Tool results
+auto-compressed via RTK (47 filters: git-diff, grep, test-jest, terraform-plan,
+docker-logs, etc.) — 20-40% token savings. Disable per-request with
+`X-Omniroute-Rtk: off` header.
+
+### Reasoning / thinking
+
+Anthropic extended thinking and OpenAI Responses reasoning blocks are forwarded
+verbatim. Cached automatically via reasoning cache.
+
+### Errors
+
+- `401` → invalid API key
+- `400 invalid_model` → model not in registry; check `/v1/models`
+- `503 circuit_open` → provider circuit breaker tripped; retry later or use combo
+- `429 rate_limited` → honor `Retry-After`; consider using a combo for auto-fallback
+
+## Image generation
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+### Endpoints
+
+- `POST $OMNIROUTE_URL/v1/images/generations` — Text-to-image
+- `POST $OMNIROUTE_URL/v1/images/edits` — Image edit (mask)
+- `POST $OMNIROUTE_URL/v1/images/variations` — Variations
+
+### Discover
+
+```bash
+curl $OMNIROUTE_URL/v1/models/image | jq '.data[]'
+```
+
+Returns `{ id, owned_by, sizes:[...], capabilities:[...] }` per model.
+
+### Generate example
+
+```bash
+curl -X POST $OMNIROUTE_URL/v1/images/generations \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "model": "dall-e-3",
+ "prompt": "a red bicycle on a wet street, photoreal",
+ "n": 1,
+ "size": "1024x1024",
+ "response_format": "b64_json"
+ }'
+```
+
+Response: `{ created, data: [{ url? or b64_json, revised_prompt }] }`
+
+### Errors
+
+- `400 invalid_size` → not supported by this model; check `/v1/models/image`
+- `400 content_policy_violation` → blocked by provider safety
+- `503` → provider unavailable; try another model in `/v1/models/image`
+
+## Text-to-speech
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+### Endpoint
+
+- `POST $OMNIROUTE_URL/v1/audio/speech` — returns binary audio (mp3/opus/wav/flac)
+
+### Discover
+
+```bash
+curl $OMNIROUTE_URL/v1/models/tts | jq '.data[]'
+```
+
+Each entry includes `voices:[...]` for the available voice names per provider.
+
+### Example
+
+```bash
+curl -X POST $OMNIROUTE_URL/v1/audio/speech \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "model": "tts-1",
+ "input": "Hello from OmniRoute.",
+ "voice": "alloy",
+ "response_format": "mp3"
+ }' --output speech.mp3
+```
+
+### Voices
+
+Voice names vary by provider. Check `/v1/models/tts` — each entry has `voices:[...]`.
+Common OpenAI voices: `alloy`, `echo`, `fable`, `onyx`, `nova`, `shimmer`.
+
+### Errors
+
+- `400 invalid_voice` → voice not supported by this model
+- `400 input_too_long` → input exceeds model character limit
+- `503` → provider unavailable; try another model in `/v1/models/tts`
+
+## Speech-to-text
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+### Endpoints
+
+- `POST $OMNIROUTE_URL/v1/audio/transcriptions` — multipart upload, returns text
+- `POST $OMNIROUTE_URL/v1/audio/translations` — transcribe + translate to English
+
+### Discover
+
+```bash
+curl $OMNIROUTE_URL/v1/models/stt | jq '.data[]'
+```
+
+### Example
+
+```bash
+curl -X POST $OMNIROUTE_URL/v1/audio/transcriptions \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -F "model=whisper-1" \
+ -F "response_format=verbose_json"
+```
+
+Response: `{ text, language, duration, segments?:[{ start, end, text }] }`
+
+### Supported formats
+
+Audio: `mp3`, `mp4`, `mpeg`, `mpga`, `m4a`, `wav`, `webm`.
+Response formats: `json`, `text`, `srt`, `verbose_json`, `vtt`.
+
+### Errors
+
+- `400 invalid_file_format` → unsupported audio format
+- `400 file_too_large` → exceeds provider limit (usually 25MB)
+- `503` → provider unavailable; try another model in `/v1/models/stt`
+
+## Embeddings
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+### Endpoint
+
+- `POST $OMNIROUTE_URL/v1/embeddings`
+
+### Discover
+
+```bash
+curl $OMNIROUTE_URL/v1/models/embedding | jq '.data[]'
+```
+
+Each entry: `{ id, owned_by, dimensions, max_input_tokens }`.
+
+### Example
+
+```bash
+curl -X POST $OMNIROUTE_URL/v1/embeddings \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "model": "text-embedding-3-large",
+ "input": ["first text", "second text"],
+ "encoding_format": "float"
+ }'
+```
+
+Response: `{ data:[{ embedding:[...], index }], usage:{ prompt_tokens, total_tokens } }`
+
+### Batch input
+
+`input` accepts a string or array of strings (up to provider batch limit, typically 2048 items).
+
+### Errors
+
+- `400 input_too_long` → input exceeds `max_input_tokens` for this model
+- `400 invalid_encoding_format` → use `float` or `base64`
+- `503` → provider unavailable; try another model in `/v1/models/embedding`
+
+## Web search
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+### Endpoint
+
+- `POST $OMNIROUTE_URL/v1/web/search` — unified search format
+
+### Discover
+
+```bash
+curl $OMNIROUTE_URL/v1/models/web | jq '.data[] | select(.kind == "webSearch")'
+```
+
+### Example
+
+```bash
+curl -X POST $OMNIROUTE_URL/v1/web/search \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "model": "tavily/search",
+ "query": "OmniRoute github latest release",
+ "max_results": 5,
+ "include_answer": true
+ }'
+```
+
+Response: `{ answer?, results:[{ url, title, content, score }] }`
+
+### Parameters
+
+| Field | Type | Description |
+| ---------------- | ------- | ------------------------------------ |
+| `model` | string | Provider model from `/v1/models/web` |
+| `query` | string | Search query |
+| `max_results` | number | Max results (default: 5) |
+| `include_answer` | boolean | Include AI-synthesized answer |
+| `search_depth` | string | `basic` or `advanced` (Tavily) |
+
+### Errors
+
+- `400 query_too_long` → shorten the search query
+- `503` → provider unavailable; try another model in `/v1/models/web`
+
+## Web fetch
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+### Endpoint
+
+- `POST $OMNIROUTE_URL/v1/web/fetch`
+
+### Discover
+
+```bash
+curl $OMNIROUTE_URL/v1/models/web | jq '.data[] | select(.kind == "webFetch")'
+```
+
+### Example
+
+```bash
+curl -X POST $OMNIROUTE_URL/v1/web/fetch \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "model": "jina/reader",
+ "url": "https://anthropic.com",
+ "format": "markdown"
+ }'
+```
+
+Response: `{ url, title, markdown, links?:[...], images?:[...] }`
+
+### Parameters
+
+| Field | Type | Description |
+| -------- | ------ | ----------------------------------------------------------------------- |
+| `model` | string | Provider from `/v1/models/web` (e.g. `jina/reader`, `firecrawl/scrape`) |
+| `url` | string | URL to fetch |
+| `format` | string | `markdown` (default), `html`, `text` |
+
+### Errors
+
+- `400 invalid_url` → URL must be http/https
+- `403 blocked` → provider blocked by target site; try a different model
+- `503` → provider unavailable; try another model in `/v1/models/web`
+<!-- skill:custom-end -->
diff --git a/skills/omni-mcp/SKILL.md b/skills/omni-mcp/SKILL.md
new file mode 100644
index 0000000000..702bc3382d
--- /dev/null
+++ b/skills/omni-mcp/SKILL.md
@@ -0,0 +1,75 @@
+---
+name: omni-mcp
+description: OmniRoute exposes a built-in MCP (Model Context Protocol) server with 37 tools (chat, embeddings, memory CRUD, skills, providers, routing, audit) over SSE/stdio/HTTP transports. Use when the user wants to add OmniRoute as an MCP server in Claude Desktop, Cursor, Cline, or any MCP-compatible client.
+---
+
+<!-- skill:custom-start -->
+<!-- Migrated from skills/omniroute-mcp/SKILL.md (preserved curated content) -->
+
+# OmniRoute — MCP Server
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+## Transports
+
+- **stdio** — local IPC, for Claude Desktop / VS Code extensions
+- **SSE** — `GET $OMNIROUTE_URL/api/mcp/sse`
+- **Streamable HTTP** — `POST $OMNIROUTE_URL/api/mcp/stream`
+
+## Claude Desktop config
+
+Add to `~/Library/Application Support/Claude/claude_desktop_config.json`:
+
+```json
+{
+ "mcpServers": {
+ "omniroute": {
+ "command": "npx",
+ "args": ["-y", "omniroute", "--mcp"],
+ "env": { "OMNIROUTE_KEY": "sk-..." }
+ }
+ }
+}
+```
+
+## Cursor / VS Code config
+
+```json
+{
+ "mcp": {
+ "servers": {
+ "omniroute": {
+ "url": "http://localhost:20128/api/mcp/sse",
+ "headers": { "Authorization": "Bearer sk-..." }
+ }
+ }
+ }
+}
+```
+
+## Available tools (37 total)
+
+| Scope | Tools |
+| --------- | -------------------------------------------------------------------------------------------------- |
+| health | `omniroute_get_health` |
+| combos | `omniroute_list_combos`, `omniroute_get_combo_metrics`, `omniroute_switch_combo` |
+| routing | `omniroute_simulate_route`, `omniroute_best_combo_for_task`, `omniroute_explain_route` |
+| providers | `omniroute_get_provider_metrics`, `omniroute_check_quota`, `omniroute_route_request` |
+| budget | `omniroute_set_budget_guard`, `omniroute_set_routing_strategy`, `omniroute_set_resilience_profile` |
+| testing | `omniroute_test_combo` |
+| memory | `memory_add`, `memory_search`, `memory_delete` |
+| skills | `skill_invoke`, `skill_list`, `skill_describe`, `skill_register` |
+| cache | `omniroute_cache_stats`, `omniroute_cache_flush` |
+| admin | `omniroute_db_health_check`, `omniroute_sync_pricing`, `omniroute_get_session_snapshot` |
+
+Full list: `GET $OMNIROUTE_URL/api/mcp/tools`
+
+## Scopes
+
+Tools are grouped into 13 scopes (chat-only, memory-readonly, full-admin, etc.).
+Pass scope name as `--scope` arg or via `X-Omniroute-Scope` header.
+
+## Reference
+
+Full docs: https://github.com/diegosouzapw/OmniRoute/blob/main/docs/frameworks/MCP-SERVER.md
+<!-- skill:custom-end -->
diff --git a/skills/omni-resilience/SKILL.md b/skills/omni-resilience/SKILL.md
new file mode 100644
index 0000000000..0da3f3830a
--- /dev/null
+++ b/skills/omni-resilience/SKILL.md
@@ -0,0 +1,118 @@
+---
+name: omni-resilience
+description: Monitor OmniRoute system health, provider circuit breakers, per-provider latency (p50/p95/p99), quota usage, and set budget guards. Use when the user wants to check if the system is healthy, debug slow providers, manage spend limits, or set up monitoring.
+---
+
+<!-- skill:custom-start -->
+<!-- Migrated from skills/omniroute-monitoring/SKILL.md (preserved curated content) -->
+
+# OmniRoute — Monitoring & Health
+
+Requires `OMNIROUTE_URL` and `OMNIROUTE_KEY`. See [entry-point SKILL](https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md) for setup.
+
+## System health
+
+```bash
+curl $OMNIROUTE_URL/api/health \
+ -H "Authorization: Bearer $OMNIROUTE_KEY"
+```
+
+Returns: uptime, memory, active connections, circuit breaker states, rate limit status, cache stats.
+
+Unauthenticated quick check:
+
+```bash
+curl $OMNIROUTE_URL/api/health
+# → {"ok":true}
+```
+
+## Provider circuit breakers
+
+Circuit breakers prevent traffic from hitting failing providers.
+
+States: `CLOSED` (normal), `OPEN` (blocked), `HALF_OPEN` (probe mode — auto-recovers).
+
+```bash
+curl $OMNIROUTE_URL/api/monitoring/health \
+ -H "Authorization: Bearer $OMNIROUTE_KEY"
+```
+
+Response includes `circuitBreakers` array with per-provider state and `resetAt` timestamp.
+
+## Per-provider metrics (p50/p95/p99)
+
+```bash
+curl $OMNIROUTE_URL/api/providers/metrics \
+ -H "Authorization: Bearer $OMNIROUTE_KEY"
+```
+
+Response shape per provider:
+
+```json
+{
+ "provider": "anthropic",
+ "requests": 1247,
+ "successRate": 0.994,
+ "latency": { "p50": 820, "p95": 2100, "p99": 3800 },
+ "circuitState": "CLOSED",
+ "tokensUsed": 2847000
+}
+```
+
+## Via MCP (if OmniRoute is your MCP server)
+
+```
+omniroute_get_health → full system health snapshot
+omniroute_get_provider_metrics → p50/p95/p99 + circuit state per provider
+omniroute_get_session_snapshot → cost, tokens, errors for current session
+omniroute_check_quota → quota balance + percent remaining + reset time
+omniroute_db_health_check → diagnose + auto-repair database drift
+```
+
+## Quota check
+
+```bash
+curl $OMNIROUTE_URL/api/quota \
+ -H "Authorization: Bearer $OMNIROUTE_KEY"
+```
+
+Returns used/total tokens and requests per provider/account, with `resetAt` timestamps.
+
+## Budget guard (spend limit)
+
+Set a session spending limit that degrades or blocks requests when hit:
+
+```bash
+curl -X POST $OMNIROUTE_URL/api/budget/guard \
+ -H "Authorization: Bearer $OMNIROUTE_KEY" \
+ -H "Content-Type: application/json" \
+ -d '{
+ "limitUsd": 5.00,
+ "action": "degrade",
+ "degradeTo": "openai/gpt-4o-mini"
+ }'
+```
+
+`action` options:
+
+- `degrade` — switch to a cheaper model when limit is hit
+- `block` — return 429 when limit is hit
+- `alert` — continue but add `X-Budget-Warning` header
+
+## MCP audit log
+
+OmniRoute logs every MCP tool call to `mcp_audit` table. Query via API:
+
+```bash
+curl "$OMNIROUTE_URL/api/mcp/status" \
+ -H "Authorization: Bearer $OMNIROUTE_KEY"
+```
+
+Returns: server status, heartbeat, recent audit activity summary.
+
+## Errors
+
+- `503` on health endpoint → OmniRoute is starting up; retry in 5s
+- Circuit breaker `OPEN` → provider is temporarily blocked; check `resetAt` to know when it auto-recovers
+- `429 budget_exceeded` → budget guard limit reached; raise limit or wait for reset
+<!-- skill:custom-end -->