Skip to content

Business operating system for Claude Code — 57 skills, 21 agents, smart daemon. Unified inbox (WhatsApp/Email/Slack/Telegram), autonomous PR merge, full-AWS monitoring, revenue (Stripe+RevenueCat), e-commerce (Shopify), marketing (Klaviyo/Meta/GA4), voice (Bland/ElevenLabs), APM (Datadog/NewRelic/OTEL), YOLO mode.

fix: Notion integration fixes + bump to v1.3.0

auroracapitalfce90cee6d28
diff --git a/claude-ops/package.json b/claude-ops/package.json
index 2975da79..d9ec488e 100644
--- a/claude-ops/package.json
+++ b/claude-ops/package.json
@@ -1,6 +1,6 @@
{
"name": "claude-ops-bin",
- "version": "1.2.0",
+ "version": "1.3.0",
"private": true,
"description": "Runtime dependencies for claude-ops bin scripts (ops-telegram-autolink.mjs, ops-slack-autolink.mjs). Installed separately from telegram-server/ so the .mjs scripts resolve modules from claude-ops/node_modules regardless of invocation cwd.",
"type": "module",
diff --git a/claude-ops/skills/ops-comms/SKILL.md b/claude-ops/skills/ops-comms/SKILL.md
index 195d35b4..9d44df50 100644
--- a/claude-ops/skills/ops-comms/SKILL.md
+++ b/claude-ops/skills/ops-comms/SKILL.md
@@ -173,11 +173,13 @@ Fall back to: `telegram-cli --exec "dialog_list" 2>/dev/null || echo "Telegram M
`${CLAUDE_PLUGIN_ROOT}/bin/ops-discord read "<CHANNEL_ID>" --limit 20 --json` — requires `DISCORD_BOT_TOKEN` (or credential-store `discord/bot-token`). Fall back to `bin/ops-discord channels --json` if the user doesn't know the channel ID and `DISCORD_GUILD_ID` is set.
**Notion:**
-Use `mcp__claude_ai_Notion__notion-search` with the user's query (or `query: "recent"` for general browsing). For each result:
+Use `mcp__claude_ai_Notion__notion-search` with the user's query (or `query: ""` sorted by `last_edited_time` for general browsing). For each result:
- Fetch full page content with `mcp__claude_ai_Notion__notion-fetch` using the page URL/ID from search results
- Get comments with `mcp__claude_ai_Notion__notion-get-comments`
- Show page title, database name, last editor, and recent comments
+**Notion API fallback:** If MCP tools fail and `NOTION_API_KEY` is set, use `curl -s -H "Authorization: Bearer $NOTION_API_KEY" -H "Notion-Version: 2022-06-28" -X POST https://api.notion.com/v1/search -d '{"query":"<QUERY>","page_size":10}'`
+
### Notion comment/reply
Use `mcp__claude_ai_Notion__notion-create-comment` with the page ID to reply to a comment thread. For creating new pages in a database, use `mcp__claude_ai_Notion__notion-create-pages`.
diff --git a/claude-ops/skills/ops-go/SKILL.md b/claude-ops/skills/ops-go/SKILL.md
index 728e0fbd..3664992d 100644
--- a/claude-ops/skills/ops-go/SKILL.md
+++ b/claude-ops/skills/ops-go/SKILL.md
@@ -173,7 +173,7 @@ After the briefing, use **batched AskUserQuestion calls** (max 4 options each) f
For Slack counts: if the pre-gathered data shows `"count": -1`, use `mcp__claude_ai_Slack__slack_search_public_and_private` with query `in:channel` (NOT `is:unread` — scan full recent activity) to get actual message counts. Do this as a parallel tool call while analyzing other data.
-For Notion counts: if `NOTION_MCP_ENABLED=true` and pre-gathered data shows Notion as available, use `mcp__claude_ai_Notion__notion-search` with `query: "comment"` filtered to the last 7 days to surface recent comments and mentions. Count items needing response.
+For Notion counts: if `NOTION_MCP_ENABLED=true` and pre-gathered data shows Notion as available, use `mcp__claude_ai_Notion__notion-search` with `query: ""` sorted by `last_edited_time` descending to surface recently active pages. Then call `mcp__claude_ai_Notion__notion-get-comments` on the top results to find comments needing response. Note: Notion search does not support date range filters — sort by recency and limit to the first 10-20 results instead.
---
diff --git a/claude-ops/skills/ops-inbox/SKILL.md b/claude-ops/skills/ops-inbox/SKILL.md
index a3cd6726..d0c09038 100644
--- a/claude-ops/skills/ops-inbox/SKILL.md
+++ b/claude-ops/skills/ops-inbox/SKILL.md
@@ -29,6 +29,7 @@ allowed-tools:
- mcp__claude_ai_Notion__notion-get-comments
- mcp__claude_ai_Notion__notion-create-comment
- mcp__claude_ai_Notion__notion-update-page
+ - mcp__claude_ai_Notion__notion-create-pages
effort: high
maxTurns: 60
---
@@ -467,10 +468,12 @@ Notion serves as a knowledge base and task management channel. Unlike messaging
**Phase 1 — Discover and scan:**
1. Search for recent activity using `mcp__claude_ai_Notion__notion-search`:
- - `query: "@me"` or `query: "comment"` with `query_type: "internal"` — find pages with recent mentions/comments
- - Filter by `created_date_range` to last 7 days for relevance
+ - Use broad queries like `query: ""` (empty string returns recent pages) or topic-specific terms
+ - Use `filter: {"property": "object", "value": "page"}` to limit to pages (not databases)
+ - Sort by `last_edited_time` descending to surface recent activity
+ - Note: Notion search is full-text over titles/content — it does NOT support mention-based queries or date range filters
2. For each result, fetch full content: `mcp__claude_ai_Notion__notion-fetch` with the page URL/ID
-3. Get comments on active pages: `mcp__claude_ai_Notion__notion-get-comments` with the page ID
+3. Get comments on active pages: `mcp__claude_ai_Notion__notion-get-comments` with the page ID — scan comment authors and timestamps to determine which need replies
**Phase 2 — Classify:**
@@ -519,6 +522,15 @@ Use `AskUserQuestion` for each NEEDS REPLY item.
- Use `mcp__claude_ai_Notion__notion-update-page` to change status, add notes
- Only update properties the user explicitly approves
+**API fallback (when MCP is down):**
+If Notion MCP tools fail or are unavailable but `NOTION_API_KEY` is set, fall back to direct API:
+```bash
+curl -s -H "Authorization: Bearer $NOTION_API_KEY" -H "Notion-Version: 2022-06-28" \
+ -H "Content-Type: application/json" \
+ -X POST https://api.notion.com/v1/search \
+ -d '{"sort":{"direction":"descending","timestamp":"last_edited_time"},"page_size":10}'
+```
+
If `NOTION_MCP_ENABLED` is not set or Notion MCP tools are unavailable, report: "Notion not configured — set NOTION_MCP_ENABLED=true and add Notion integration via claude.ai or self-hosted MCP".
### Discord (v1 — REST channel scan)
diff --git a/claude-ops/skills/setup/SKILL.md b/claude-ops/skills/setup/SKILL.md
index b726fe49..41979f4b 100644
--- a/claude-ops/skills/setup/SKILL.md
+++ b/claude-ops/skills/setup/SKILL.md
@@ -1188,9 +1188,13 @@ For claude.ai integration:
Test the integration (run in background):
```bash
-# For claude.ai: integration auto-detected
-# For self-hosted: verify MCP server responds
-echo '{"test": true}' | timeout 10 npx -y @notionhq/notion-mcp-server 2>/dev/null && echo "OK" || echo "FAIL"
+# For claude.ai: integration auto-detected — test via MCP tool call
+# For self-hosted: verify API key works
+if [ -n "$NOTION_API_KEY" ]; then
+ curl -s -o /dev/null -w "%{http_code}" -H "Authorization: Bearer $NOTION_API_KEY" -H "Notion-Version: 2022-06-28" https://api.notion.com/v1/users/me | grep -q "200" && echo "OK" || echo "FAIL"
+else
+ echo "OK — claude.ai integration (verify via MCP tool call after restart)"
+fi
```
#### Finalize
@@ -1204,8 +1208,6 @@ echo '{"test": true}' | timeout 10 npx -y @notionhq/notion-mcp-server 2>/dev/nul
### 3f — Calendar
-> **Note:** Sections after this were originally labeled 3f–3m. The insertion of Notion (3e) shifted Calendar to 3f. Downstream section headers retain their original labels to avoid breaking cross-references.
-
Calendar isn't a messaging channel, but every other ops skill (briefings, `/ops-next`, `/ops-go`) benefits massively from knowing the user's schedule — meetings blocking deep work, deploy windows, travel days. The wizard wires it up the same way as email: `gog calendar` primary, Google Calendar MCP connector fallback.
#### Preferred: `gog calendar`
@@ -1265,11 +1267,11 @@ Downstream skills (`/ops-go`, `/ops-next`, `/ops-fires`) read `channels.calendar
> **Deep-dive:** see `${CLAUDE_PLUGIN_ROOT}/skills/ops-go/SKILL.md` for full operational instructions, CLI reference, and troubleshooting for this integration (calendar context feeds `/ops:go` briefings). The setup agent can load that file directly when it needs more depth than this wizard provides.
-### 3f — Doppler (secrets management)
+### 3g — Doppler (secrets management)
Doppler is a secrets manager that injects environment variables at runtime. When configured, all ops skills can query secrets via `doppler secrets get` instead of reading from dotfiles or keychain. The wizard checks presence, auth status, and default project context.
-#### Step 3f.1 — Presence
+#### Step 3g.1 — Presence
```bash
command -v doppler
@@ -1297,7 +1299,7 @@ Where the OS-specific command is:
Run the chosen command in the background, capture stdout/stderr, and report success/failure. If the user skips, record `secrets_manager: "none"` in `$PREFS_PATH` and end this sub-flow.
-#### Step 3f.2 — Auth status
+#### Step 3g.2 — Auth status
Run:
@@ -1319,7 +1321,7 @@ Run `doppler login` via Bash tool with `run_in_background: true` (it opens a bro
Never display the name or email unless they came from `doppler me` output in this session.
-#### Step 3f.3 — Project context
+#### Step 3g.3 — Project context
If authenticated, list available projects:
@@ -1370,7 +1372,7 @@ Print confirmation:
✓ Doppler default context set: <project>/<config>
```
-#### Step 3f.4 — Document for agents
+#### Step 3g.4 — Document for agents
Print this note so it's visible in the session:
@@ -1388,11 +1390,11 @@ Individual skills can override with --project / --config flags.
> **Deep-dive:** no dedicated skill ships with Doppler — see `${CLAUDE_PLUGIN_ROOT}/docs/memories-system.md` (Runtime Context section) for how downstream skills consume the `secrets_manager` / `doppler.*` values from `$PREFS_PATH` and resolve `doppler:KEY_NAME` references at runtime. The setup agent can load that file directly when it needs more depth than this wizard provides.
-### 3g — Password Manager (credential vault)
+### 3h — Password Manager (credential vault)
Ops agents frequently need to look up credentials (API keys, database passwords, service tokens) on your behalf. This step wires up a password manager so those queries can be automated via a standard command template stored in `$PREFS_PATH`.
-#### Step 3g.1 — Auto-detect installed managers
+#### Step 3h.1 — Auto-detect installed managers
Run these in parallel:
@@ -1405,7 +1407,7 @@ security find-generic-password -s "test" 2>&1 | head -1 # macOS Keyc
Parse each result to classify as `authenticated`, `needs_unlock`, `not_installed`, or `available` (Keychain is always `available`).
-#### Step 3g.2 — Present findings
+#### Step 3h.2 — Present findings
Show only what was detected via `AskUserQuestion`. **Max 4 options per call.** Since macOS Keychain and Skip are always shown, you have room for at most 2 detected managers per call. If all 3 CLIs (1Password, Dashlane, Bitwarden) are installed, batch into two calls:
@@ -1434,7 +1436,7 @@ Call 2:
Never show managers that aren't installed. Always show macOS Keychain and Skip. If none of the CLIs are installed, skip straight to showing just `[macOS Keychain — always available]` and `[Skip]`.
-#### Step 3g.3 — Configure selected manager
+#### Step 3h.3 — Configure selected manager
**1Password (`op`):**
@@ -1506,7 +1508,7 @@ Never show managers that aren't installed. Always show macOS Keychain and Skip.
security find-generic-password -s "{{name}}" -w
```
-#### Step 3g.4 — Write to preferences
+#### Step 3h.4 — Write to preferences
After the user selects and configures a manager, write to `$PREFS_PATH`:
@@ -1534,7 +1536,7 @@ Merge with the existing file (`jq '. + { ... }'`) — never overwrite. Example f
If the user picks Skip, write `"password_manager": "none"` so subsequent runs don't re-prompt unless the user explicitly runs `/ops:setup vault`.
-#### Step 3g.5 — Document for agents
+#### Step 3h.5 — Document for agents
After saving, print this note once:
@@ -1570,15 +1572,15 @@ Omit this line entirely if `password_manager` is `"none"` or unset.
#### Invocation shortcut
-Add to the shortcuts table: `vault`, `password-manager`, `pm` → Step 3g
+Add to the shortcuts table: `vault`, `password-manager`, `pm` → Step 3h
> **Deep-dive:** no dedicated skill ships with the password manager integration — see `${CLAUDE_PLUGIN_ROOT}/docs/memories-system.md` (Runtime Context section) for how downstream skills resolve `password_manager` + related vault references from `$PREFS_PATH`. Privacy-and-security guidance lives in this SKILL.md (keychain-only storage of API hashes/session strings, `umask 077` for bridge files). The setup agent can load that file directly when it needs more depth than this wizard provides.
---
-### 3h — Ecommerce (Shopify + dynamic partners)
+### 3i — Ecommerce (Shopify + dynamic partners)
-#### Step 3h.1 — Auto-scan for existing Shopify credentials
+#### Step 3i.1 — Auto-scan for existing Shopify credentials
**Before asking for anything**, run the Universal Credential Auto-Scan for all Shopify-related vars simultaneously:
@@ -1623,16 +1625,16 @@ dcli password shopify --output json 2>/dev/null | jq -r '.[].url // empty' | gre
grep -rhE 'myshopify\.com|SHOPIFY_STORE' ~/Projects/*/.env* 2>/dev/null | grep -v '^#' | head -5
```
-**Important**: Do NOT report "No Shopify credentials found" until ALL scan sources have been checked. If tokens are missing but store URLs are found (e.g. from Chrome history or Dashlane), report: `"Found Shopify store(s): <stores>. No API token found — you'll need to create one."` and skip straight to Step 3h.3 (token) with the store URL pre-filled.
+**Important**: Do NOT report "No Shopify credentials found" until ALL scan sources have been checked. If tokens are missing but store URLs are found (e.g. from Chrome history or Dashlane), report: `"Found Shopify store(s): <stores>. No API token found — you'll need to create one."` and skip straight to Step 3i.3 (token) with the store URL pre-filled.
If both `store_url` and `admin_token` are already found, show:
```
✓ Shopify — already configured (<store_url>)
[Keep existing] [Reconfigure]
```
-If the user keeps existing, skip to Step 3h.4. If reconfiguring or no values found, continue.
+If the user keeps existing, skip to Step 3i.4. If reconfiguring or no values found, continue.
-#### Step 3h.2 — Shopify store URL
+#### Step 3i.2 — Shopify store URL
If `SHOPIFY_STORE_URL` was found in the auto-scan, present it using the Universal Credential Auto-Scan prompt format. Only ask via free text if no value was found:
```
@@ -1643,7 +1645,7 @@ Enter your Shopify store URL:
Validate the input: strip `https://`, strip trailing slash, check that the result ends with `.myshopify.com`. If invalid, ask again with a correction note.
-#### Step 3h.3 — Shopify Admin API token
+#### Step 3i.3 — Shopify Admin API token
If `SHOPIFY_ACCESS_TOKEN`, `SHOPIFY_ADMIN_TOKEN`, or `SHOPIFY_ADMIN_API_ACCESS_TOKEN` was found in the auto-scan, present it using the Universal Credential Auto-Scan prompt format with truncated display (`shpat_508b...682e`). Only ask via free text if no value was found.
@@ -1693,7 +1695,7 @@ curl -s -H "X-Shopify-Access-Token: $TOKEN" \
```
Expect a shop name string. If the response contains `errors` or `{"shop":null}`, show the error and ask the user to check the token scopes. Print `✓ Shopify — connected (<shop name>)`.
-#### Step 3h.4 — Dynamic ecommerce partners
+#### Step 3i.4 — Dynamic ecommerce partners
After Shopify is configured, ask via `AskUserQuestion` (free text):
```
@@ -1769,7 +1771,7 @@ For any partner not in this table, always web search for current auth docs befor
---
-### 3i — Marketing (Klaviyo, Meta Ads, GA4, Search Console)
+### 3j — Marketing (Klaviyo, Meta Ads, GA4, Search Console)
**Before showing the service selector**, run the Universal Credential Auto-Scan for all marketing vars simultaneously:
@@ -1871,7 +1873,7 @@ Write to `$PREFS_PATH` (merge):
}
```
-Same Doppler-reference pattern as Step 3h — prefer `doppler:KEY_NAME` over raw tokens when Doppler is configured.
+Same Doppler-reference pattern as Step 3i — prefer `doppler:KEY_NAME` over raw tokens when Doppler is configured.
#### Dynamic marketing partners
@@ -1883,7 +1885,7 @@ Any other marketing tools you'd like to connect?
Type names separated by commas, or leave blank to skip.
```
-If the user provides partner names, apply the same dynamic partner loop as Step 3h.4 — for each partner:
+If the user provides partner names, apply the same dynamic partner loop as Step 3i.4 — for each partner:
1. **Research credentials** via web search: `"<partner name> API authentication developer docs 2025"`
2. **Ask for credentials** via `AskUserQuestion` with instructions sourced from the docs
@@ -1923,7 +1925,7 @@ For any partner not in this table, always web search for current auth docs befor
---
-### 3j — Voice (Bland AI, ElevenLabs, Groq)
+### 3k — Voice (Bland AI, ElevenLabs, Groq)
**Before showing the service selector**, run the Universal Credential Auto-Scan for all voice vars simultaneously:
@@ -2022,7 +2024,7 @@ Same Doppler-reference pattern — prefer `doppler:KEY_NAME` over raw tokens whe
---
-### 3k — Revenue (Stripe + RevenueCat)
+### 3l — Revenue (Stripe + RevenueCat)
**Before showing the service selector**, run the Universal Credential Auto-Scan for all revenue vars simultaneously (Rule 4 — background these):
@@ -2173,7 +2175,7 @@ Prefer a Doppler reference (`doppler:STRIPE_SECRET_KEY`, `doppler:REVENUECAT_API
---
-### 3m — Notifications (fires-watcher sinks)
+### 3n — Notifications (fires-watcher sinks)
Sets up push notifications for CRITICAL/HIGH fires so the user stops having to poll `/ops:fires` manually. Gated behind the `fires-watcher` daemon service (disabled by default).
@@ -2279,7 +2281,7 @@ Expect `{"status": "ok", ...}` within 60 seconds.
> **Deep-dive:** see `${CLAUDE_PLUGIN_ROOT}/docs/notifications.md`, `${CLAUDE_PLUGIN_ROOT}/scripts/ops-fires-watcher.sh`, and `${CLAUDE_PLUGIN_ROOT}/scripts/ops-notify.sh` for sink priority rationale, debounce rules, and a troubleshooting walk-through.
-### 3l — Discord (webhook + optional bot)
+### 3m — Discord (webhook + optional bot)
Discord is a v1 integration — webhook-based send + REST channel reads. DM + gateway support are deferred to a v2 issue. The send-side webhook also supplies the Discord notification sink consumed by `scripts/ops-notify.sh`, so configuring it here covers both `/ops:comms discord send` and ops-fires alerts.
@@ -2746,12 +2748,13 @@ If `$ARGUMENTS` contains a specific section name, jump straight to that section:
| `whatsapp`, `wacli`, `whatsapp-doctor` | Step 3b |
| `email` | Step 3c |
| `slack` | Step 3d |
-| `calendar`, `cal` | Step 3e |
-| `doppler`, `secrets` | Step 3f |
-| `vault`, `password-manager`, `pm` | Step 3g |
-| `ecom`, `shopify`, `store` | Step 3h |
-| `marketing`, `klaviyo`, `ads`, `meta`, `ga4` | Step 3i |
-| `voice`, `bland`, `elevenlabs`, `tts` | Step 3j |
+| `notion` | Step 3e |
+| `calendar`, `cal` | Step 3f |
+| `doppler`, `secrets` | Step 3g |
+| `vault`, `password-manager`, `pm` | Step 3h |
+| `ecom`, `shopify`, `store` | Step 3i |
+| `marketing`, `klaviyo`, `ads`, `meta`, `ga4` | Step 3j |
+| `voice`, `bland`, `elevenlabs`, `tts` | Step 3k |
| `mcp` | Step 4 |
| `registry`, `projects` | Step 5 |
| `daemon`, `background` | Step 5b |