A local-first multi-agent app development platform, built on PenguinHarness
feat(skills): penguin-orchestration — drive the harness from inside an agent (#463)
Yaowei Zhengf8142340be92
diff --git a/changelog/unreleased/2026-08-25-penguin-orchestration-skill.md b/changelog/unreleased/2026-08-25-penguin-orchestration-skill.mdnew file mode 100644index 00000000..9321a6ba--- /dev/null+++ b/changelog/unreleased/2026-08-25-penguin-orchestration-skill.md@@ -0,0 +1,16 @@+# New built-in skill: penguin-orchestration++- **Date:** 2026-08-25+- **Type:** feature+- **Scope:** `skills`, `docs`+- **PR:** [#463](https://github.com/Prism-Shadow/penguin-harness/pull/463)++[中文版](2026-08-25-penguin-orchestration-skill.zh.md)++Added `penguin-orchestration` (v1) to the skill library's AI App Development group, right after `penguin-cli`. It teaches an agent running inside PenguinHarness to drive the harness itself over the `penguin` CLI against its own local server: list and create agents, start sessions and steer them mid-flight, read transcripts, and query costs and scheduled tasks. The skill documents the server-backed CLI surface introduced on branch `feat/cli-on-server` — the injected `PENGUIN_API_URL` / `PENGUIN_API_TOKEN` / project, agent and session id environment variables inside agent sessions, and the lock-file attach or auto-start path outside — so it ships together with that rework and must merge after it lands.++## Details++- Recipes: summarize yesterday's conversations (session ids embed their creation timestamp, `session-YYYY-MM-DD-HH-mm-ss-<8hex>`), query the last 7 days of cost (`penguin cost --days 7 --by agent` and variants), create an agent and talk to it (`penguin agent create` then `penguin run --agent-id`), inspect scheduled tasks (`penguin schedule ls`) and create one by writing a schedule TOML with file tools, and run a CLI-driven conversation in the background — `exec_command` with `run_in_background` for a completion report, or `penguin run --background` for a server-side run that survives the caller — steering it mid-flight with `penguin input <session_id> -m ... --no-wait`.+- Cautions cover the one-active-task-per-session rule, approval modes for unattended sessions (`allow-all` / `read-only`; `always-ask` hangs), runaway self-messaging loops, project-level cost attribution, and the standing ban on hand-editing `.project_config.toml` / `.vault.toml`.+- Registered in `SKILL_GROUPS`, the package README table and the bilingual docs skill tables; hand-drawn `icon.svg` (a conductor node fanning out to two nodes). No `preinstall` marker, so new `default_agent`s get it like the rest of the library.diff --git a/changelog/unreleased/2026-08-25-penguin-orchestration-skill.zh.md b/changelog/unreleased/2026-08-25-penguin-orchestration-skill.zh.mdnew file mode 100644index 00000000..a56b1083--- /dev/null+++ b/changelog/unreleased/2026-08-25-penguin-orchestration-skill.zh.md@@ -0,0 +1,16 @@+# 新增内置技能:penguin-orchestration++- **Date:** 2026-08-25+- **Type:** feature+- **Scope:** `skills`, `docs`+- **PR:** [#463](https://github.com/Prism-Shadow/penguin-harness/pull/463)++[English](2026-08-25-penguin-orchestration-skill.md)++技能库 AI App Development 组新增 `penguin-orchestration`(v1),排在 `penguin-cli` 之后。它教运行在 PenguinHarness 内的 Agent 经 `penguin` CLI 驱动承载自己的那台本地服务:列出与创建 Agent、发起会话并中途改向、读取会话记录、查询成本与定时任务。技能文档记载的是 `feat/cli-on-server` 分支引入的服务端化 CLI 命令面——Agent 会话内注入的 `PENGUIN_API_URL` / `PENGUIN_API_TOKEN` 及项目、Agent、会话 id 环境变量,会话外经锁文件附着或自动拉起——因此与该改造同行,须在其落地后合并。++## 细节++- 配方:总结昨天的对话(会话 id 内嵌创建时间戳 `session-YYYY-MM-DD-HH-mm-ss-<8hex>`)、查询近 7 天成本(`penguin cost --days 7 --by agent` 及各变体)、创建 Agent 并与之对话(`penguin agent create` 后接 `penguin run --agent-id`)、查看定时任务(`penguin schedule ls`)并用文件工具写 schedule TOML 新建,以及在后台跑一场 CLI 驱动的对话——`exec_command` 带 `run_in_background` 可收到完成回报,`penguin run --background` 则由服务端持续执行、不随发起者结束——中途用 `penguin input <session_id> -m ... --no-wait` 改向。+- 注意事项覆盖每会话单活跃任务规则、无人值守会话的审批模式(`allow-all` / `read-only`;`always-ask` 会挂起等待)、自我发消息的失控循环、成本记入同一 Project,以及一贯的禁止手改 `.project_config.toml` / `.vault.toml`。+- 已登记进 `SKILL_GROUPS`、包 README 表格与双语文档技能表;手绘 `icon.svg`(指挥节点扇出到两个节点)。未设 `preinstall` 标记,新建 `default_agent` 与库内其余技能一样预装它。diff --git a/packages/docs/content/skills.en.md b/packages/docs/content/skills.en.mdindex 017cd1de..33c2b496 100644--- a/packages/docs/content/skills.en.md+++ b/packages/docs/content/skills.en.md@@ -68,6 +68,7 @@ The built-in Skills, by group (the group manifest is `SKILL_GROUPS` in `packages| | `remote-claude-code` | Run Claude Code on a remote host over SSH — a persistent expect session, headless `-p` with the stdin fix, a tmux-driven interactive TUI (one capture-verified keystroke at a time, user messages relayed verbatim) and multi-turn continuity (not preinstalled: install from the library when needed) || AI App Development | `penguin-sdk` | Build agent, AI and RAG applications on the SDK — application code, not Agent State configuration: the createSession/run streaming loop, CLI-wrapped user tools, plus a complete retrieval recipe with chunk-revealing citations || | `penguin-cli` | Manage model API keys, default models and per-agent Vault secrets with the penguin CLI |+| | `penguin-orchestration` | Drive PenguinHarness itself from a shell: list and create agents and sessions, send and steer messages mid-flight, and query costs and scheduled tasks via the penguin CLI || | `agenthub-models` | Call model APIs through `@prismshadow/agenthub`: streaming text, image generation, speech synthesis and embeddings || | `vllm` | Deploy and serve LLMs with vLLM behind an OpenAI-compatible endpoint, with tool calling enabled for agent workloads || | `ollama` | Deploy and serve local models with Ollama: pull and run them, then expose the OpenAI-compatible endpoint to apps and agents |diff --git a/packages/docs/content/skills.zh.md b/packages/docs/content/skills.zh.mdindex 6d1feeb8..79edc87b 100644--- a/packages/docs/content/skills.zh.md+++ b/packages/docs/content/skills.zh.md@@ -68,6 +68,7 @@ Skill 库以 npm 包 `@prismshadow/penguin-skills` 发布,tarball 直接携带| | `remote-claude-code` | 通过 SSH 在远程主机上驱动 Claude Code:expect 持久会话、headless `-p` 的 stdin 修正、tmux 驱动的交互 TUI(逐个按键并截屏确认,用户消息原样转发)与多轮续接(不预装,按需从技能库安装) || AI 应用开发 | `penguin-sdk` | 基于 SDK 构建智能体应用、AI 与 RAG 应用(写应用代码,而非配置 Agent State):createSession/run 流式循环、以 CLI 形式接入用户已有工具,外加带可溯源引用的完整检索配方 || | `penguin-cli` | 用 penguin CLI 管理模型 API Key、默认模型与各 Agent 的 Vault 密钥 |+| | `penguin-orchestration` | 从 shell 驱动 PenguinHarness 自身:用 penguin CLI 列出与创建 Agent 和会话、发送并中途改向消息、查询成本与定时任务 || | `agenthub-models` | 经 `@prismshadow/agenthub` 调用模型 API:流式文本、图像生成、语音合成与 Embedding || | `vllm` | 用 vLLM 部署与服务 LLM,提供 OpenAI 兼容端点,并为 Agent 负载启用工具调用 || | `ollama` | 用 Ollama 部署与运行本地模型,把 OpenAI 兼容端点接入应用与 Agent |diff --git a/packages/server/test/skills.test.ts b/packages/server/test/skills.test.tsindex 791e2e37..0f3a6691 100644--- a/packages/server/test/skills.test.ts+++ b/packages/server/test/skills.test.ts@@ -90,6 +90,7 @@ describe("skills api", () => {expect(body.groups[2]!.skills.map((s) => s.name)).toEqual(["penguin-sdk","penguin-cli",+ "penguin-orchestration","agenthub-models","vllm","ollama",diff --git a/packages/skills/README.md b/packages/skills/README.mdindex 501b4e7b..ebf526a3 100644--- a/packages/skills/README.md+++ b/packages/skills/README.md@@ -10,7 +10,7 @@ Included skills, in the order of the `SKILL_GROUPS` manifest in `src/index.ts` (| --- | --- || Office Productivity | `data-analysis`, `firecrawl`, `bento-slides`, `humanizer` || Software Development | `web-design`, `software-engineering`, `remote-claude-code` |-| AI App Development | `penguin-sdk`, `penguin-cli`, `agenthub-models`, `vllm`, `ollama`, `llamafactory`, `skill-porting` |+| AI App Development | `penguin-sdk`, `penguin-cli`, `penguin-orchestration`, `agenthub-models`, `vllm`, `ollama`, `llamafactory`, `skill-porting` || Agent Tuning | `agent-initialization`, `benchmark-design`, `agent-evaluation`, `agent-optimization` |`humanizer` and `remote-claude-code` carry `preinstall: false`, so `default_agent` does not get them at initialization — they are installed from the library on demand.diff --git a/packages/skills/skills/penguin-orchestration/SKILL.md b/packages/skills/skills/penguin-orchestration/SKILL.mdnew file mode 100644index 00000000..40c36cec--- /dev/null+++ b/packages/skills/skills/penguin-orchestration/SKILL.md@@ -0,0 +1,161 @@+---+name: penguin-orchestration+description: Drive PenguinHarness itself from a shell — list and create agents and sessions, send and steer messages mid-flight, and query costs and scheduled tasks via the penguin CLI over the local server.+short_description: Orchestrate agents, sessions, costs and schedules with the penguin CLI.+short_description_zh: 用 penguin CLI 编排智能体、会话、成本与定时任务。+version: 1+updated: 2026-08-25T00:00:00Z+---++# Penguin Orchestration++The `penguin` CLI is a thin client of the PenguinHarness server. Inside a harness agent session it reaches the same server that is running you, so you can orchestrate the platform yourself: list and create agents, start conversations with them, steer those conversations while they run, and query costs and scheduled tasks.++## Before you start++If the user's message only invokes this skill (e.g. "use penguin-orchestration skill") without a concrete request, ask the user what they want to orchestrate. Read-only commands (`project ls`, `agent ls`, `ls`, `logs`, `cost`, `schedule ls`) are always safe; do not create agents, start sessions or send messages until the goal is clear.++## How the connection works++- **Inside a harness agent session** (you, now): every command subprocess has `PENGUIN_API_URL`, `PENGUIN_API_TOKEN`, `PENGUIN_PROJECT_ID`, `PENGUIN_AGENT_ID` and `PENGUIN_SESSION_ID` injected, so `penguin` commands automatically reach your own server with your project and agent as the defaults — no login step.+- **Outside an agent** (a human shell): the CLI attaches to the running local server via its lock file, or auto-starts one; the local `<data-root>/api-token` file (0600) authenticates it.+- You are operating the same server that runs you: sessions and agents you create appear live in the web UI, where the user sees and owns everything you spawn.+- The injected token is admin-equivalent. Act accordingly: stick to what the task requires, and prefer read-only commands until a mutation is clearly needed.++## Orient first++Before mutating anything, see what exists:++```bash+penguin project ls # projects on this server+penguin agent ls # agents in the current project+penguin ls --json # the project's sessions, with running state+```++`--json` on any listing gives machine-parseable output.++## Command surface++```+penguin run -m <msg> [--project-id <id>] [--agent-id <id>] [--workspace <path>]+ [--model-id <id> --provider <p>] [--approve <mode>] [--thinking <level>]+ [--session <session_id>] [--background] [--timeout <duration>]+ [--goal [budget]] [--json]+penguin ls [--project-id <id>] [--agent-id <id>] [--days <n>] [-a|--all] [--json]+penguin input <session_id> [-m <text>] [--timeout <duration>]+ [--project-id <id>] [--json] [--server <url>]+penguin logs <session_id> [--project-id <id>] [--tail <n>] [-f|--follow]+ [--timeout <duration>] [--json]+penguin agent ls [--project-id <id>] [--json]+penguin agent create --agent-id <id> [--name <s>] [--description <s>] [--skills <a,b>]+ [--project-id <id>] [--json]+penguin project ls [--json]+penguin cost [--days <n>] [--from <d> --to <d>] [--by date|agent|model|session]+ [--project-id <id>] [--agent-id <id>] [--json]+penguin schedule ls [--project-id <id>] [--agent-id <id>] [--json]+penguin schedule add <name> --prompt <s> --start-at <ISO|now> [--period <duration>]+ [--end-at <ISO>] [--session-id <id> | --workspace <path>+ [--model-id <id> --provider <p>]] [--disabled]+ [--project-id <id>] [--agent-id <id>]+penguin schedule update <name> [<same field flags>] [--enable|--disable]+ [--project-id <id>] [--agent-id <id>]+penguin schedule rm <name> [--project-id <id>] [--agent-id <id>]+```++- `run` starts a task and waits, rendering the conversation, unless `--background` — then it prints the new session id and exits while the server keeps running the task. `--session <session_id>` runs the task in an existing session instead of creating one; the model reference is the `--provider` + `--model-id` pair (both or neither); `--goal [budget]` runs in goal mode — the session loops until the agent declares the goal complete, with an optional spend budget.+- **Caller-context defaults.** Inside a harness agent, a session-creating `run` fills every field you leave unspecified from your own live session, per field independently: `--workspace`, the `--model-id`/`--provider` pair, `--approve` and `--thinking` inherit the caller's values — the same convention as `run_subagent` parent inheritance. Precedence: explicit flag > caller value > plain fallback (cwd, the Project default model, `allow-all`, none — used wholesale if the caller lookup fails, with a dim stderr note). So inside an agent, `penguin run -m "..."` alone typically does the right thing; pass flags only to diverge.+- `--timeout <duration>` (`30s`, `5m`, `2h`, or bare seconds) bounds the wait of a foreground `run`, an `input`, or a `logs -f`. Expiry is a soft yield, not an error: the command exits 0 while the task keeps running server-side, printing a still-running note that names the follow-up commands (`--json` prints `{sessionId, status: "running", text}` with the text so far). `--timeout 0` (also `0s`) returns immediately after delivery — the same note without collected text (`--json`: `{sessionId, status: "running"}`); on a bare poll it snapshots a running session instantly. `run --background` stays the idiomatic fire-and-forget for new tasks and rejects `--timeout`; `logs --timeout` requires `-f`.+- `input` with `-m` steers a **running** session mid-turn (the agent absorbs it as a course correction within the current task) or starts a new turn on an idle one; it waits for the reply unless a `--timeout` bounds the wait (`--timeout 0` = deliver and return at once). Bare `input <session_id>` (no `-m`) **polls**: it prints the session's most recent complete assistant text — an idempotent snapshot that skips user/thinking/tool output and never touches approvals, mirroring `input_subagent`'s empty-prompt semantics. A running session is waited on first (bounded by `--timeout`, else indefinitely); a session with no reply yet prints `(no assistant reply yet)`. `--json` reports `{sessionId, status, text}` — `idle`/`running` when polling, `completed`/`aborted`/`running` with `-m`.+- `ls` spans every agent of the project, newest first (by last active); archived sessions are left out unless `-a`/`--all` includes them, and `--days <n>` keeps only sessions last active since local midnight n−1 days ago — today counts as day 1, so `--days 2` is yesterday and today, `--days 7` this week. `logs` renders a session's transcript: `--tail <n>` for the last entries, `-f` to follow live.+- Session ids embed their creation timestamp — `session-YYYY-MM-DD-HH-mm-ss-<8hex>`. Every `<session_id>` argument takes any unique substring of an id; the 8-hex tail is the recommended short form, and an ambiguous fragment errors listing the candidates. On `input` and `logs`, `--project-id` scopes that fragment search (unnecessary with a full id).++## Recipes++### Yesterday's or this week's sessions, with their latest replies++```bash+penguin ls --days 2 --json # yesterday + today (today counts as day 1)+penguin ls --days 7 --json # this week; add -a to include archived sessions+penguin input <session_id> # one session's latest complete assistant reply+```++- `--days <n>` keeps sessions last active since local midnight n−1 days ago. For strictly-yesterday, take `--days 2` and drop today's entries client-side — ids embed the creation date and the JSON carries last-active.+- Bare `input` prints the latest reply; add `--timeout 0` to snapshot a running session instantly instead of waiting for its turn to finish.++### Summarize this week's history in a new session++A fresh session gets a fresh context window for the summary; feed it through a file, not the prompt:++```bash+penguin ls --days 7 --json # pick the sessions+penguin logs <session_id> --tail 100 # gather each transcript (widen if cut short)+# write what you gathered into a workspace file with your file tools, then:+penguin run -m "Read ./weekly-material.md and write the weekly summary to ./weekly-summary.md"+```++- **Exchange big material through workspace files.** Caller-context defaults mean the new session shares your workspace — write the gathered transcripts to `./weekly-material.md` and have the new session read it there. Pages of transcript do not belong in `-m`.+- Read the result from `./weekly-summary.md` (the foreground run also renders the reply).++### Create an agent and say hello++```bash+penguin agent create --agent-id greeter --name "Greeter" --description "Welcomes people"+penguin run --agent-id greeter -m "Hello! Introduce yourself."+```++- Agent ids must match `^[a-z][a-z0-9_]{1,63}$`: a lowercase letter first, then lowercase letters, digits and underscores — no hyphens.+- A newly created agent starts with **no skills preinstalled**: seed it at creation with `--skills a,b` (library names) — include `penguin-orchestration` itself when the new agent must drive the harness too.+- `--agent-id` switches only the agent: workspace, model, approval and thinking still inherit from your own session (caller-context defaults) — add those flags to change them too. Each `run` without `--session` opens a fresh session; reuse a session id to continue a conversation.++### Summarize each agent's costs in a new session++```bash+penguin cost --days 7 --by agent --json # who spent what this week+penguin run -m "Summarize this per-agent cost report and flag anomalies: <the JSON>"+```++- The default `penguin cost` card already carries today / last 7 days / total; `--by date|model|session` and `--from <d> --to <d>` give the other cuts, `--project-id` / `--agent-id` narrow the scope.+- A `--by agent` report is small enough to inline in `-m`; for long breakdowns (`--by session` over a busy week), use the file-exchange pattern from the weekly-summary recipe.++### Set up a scheduled task for the current agent++```bash+penguin schedule add daily-report --prompt "Summarize yesterday's conversations" \+ --start-at now --period 1d+penguin schedule ls # verify+penguin schedule update daily-report --period 12h # adjust; --enable/--disable to toggle+penguin schedule rm daily-report # remove — no confirmation prompt+```++- `--agent-id` defaults to yourself from the caller env, so this schedules the current agent. `--start-at` takes ISO 8601 or `now`; `--period` is at least 5m (`30m`/`12h`/`1d`/`7d`), omit it for a one-shot; `--end-at` bounds recurrence. Target flags are optional: `--session-id <id>` fires into that existing session, `--workspace <path>` (optionally with the `--model-id` + `--provider` pair) pins the new-session form.+- `add` creates the schedule **enabled** (`--disabled` stages it off) — deliberately diverging from the raw file, where `enabled` defaults to false. `update` is read-modify-write: unspecified fields keep their stored values, and switching the target form clears the other one.+- In `schedule ls`, read: the AGENT column (without `--agent-id` the listing spans agents), `enabled` (a disabled entry never fires), `startAt` (first firing), `period` (absent means one-shot), the target — an existing session versus a new session per firing — and `lastFiredAt`.+- The CLI writes through the schedules API, so mistakes are rejected synchronously. The TOML file stays the single source of truth — `<app_data_dir>/agents/<agent_id>/agent_state/schedule/<name>.toml`, fields mirroring the flags (`prompt`, `enabled` — false by default in the file, `start_at`, `period`, `end_at`, `session_id` / `workspace`+`provider`+`model_id`) — and remains editable with file tools; your system prompt's schedule roster lists yours. A hand edit is only validated by the periodic reconcile (roughly every 30s), with errors landing in error records rather than your terminal — prefer the CLI.++### Run a conversation in the background and steer it mid-flight++Two patterns; both leave you free while the conversation runs.++**(a) Background CLI process — you get a completion report.** Run the CLI itself as a background command: `exec_command` with `run_in_background: true` and the command++```bash+penguin run --agent-id <agent_id> -m "<long task>"+```++- The harness delivers a `[background_task_done]` report when the CLI exits — no polling needed for completion.+- Meanwhile, find the session with `penguin ls --json` (it shows as running, with the newest id) and steer it: `penguin input <session_id> -m "Focus on X; skip Y" --timeout 0` (deliver and return at once).+- Poll the latest answer with bare `penguin input <session_id> --timeout 30s` — a bounded wait that exits 0 with a still-running note when the reply is not in yet — or read the raw transcript with `penguin logs <session_id> --tail 20`.++**(b) Server-side background — survives you.** `penguin run --background --agent-id <agent_id> -m "<long task>"` prints the session id and exits; the server keeps running the task with no local process.++- Poll the latest answer with bare `penguin input <session_id> --timeout 30s`, watch live with `penguin logs <session_id> -f`, and check running state with `penguin ls --json`; steer with `penguin input <session_id> -m ...` the same way.++Prefer (a) when you stay around for the result — the completion report comes to you. Prefer (b) when the work must survive your own session ending, or when fanning out many tasks without holding a process per task. A bounded foreground run is the middle ground: `penguin run --timeout 5m -m "..."` renders up to the bound, then soft-yields with the task still running — pick up the answer later with a bare `penguin input <session_id>`.++## Cautions++- **One active task per session.** `penguin input` at a busy session steers the running task rather than starting a second one; a new task sent at a busy session waits its turn. For parallel work, start parallel sessions.+- **Unattended sessions must not need a human.** A spawned session inherits your approval mode (`allow-all` when there is no caller to inherit from); if you yourself run under `always-ask`, pass `--approve allow-all` (trusted work) or `--approve read-only` explicitly — an unattended `always-ask` session hangs waiting for approval in the web UI.+- **No runaway loops.** An agent that messages itself — directly, through a chain of agents, or through a schedule aimed back at its own session — keeps spending until someone stops it. Make every automated conversation terminate.+- **Spawned work bills the project.** Everything you start lands in the same project's usage (`penguin cost` shows it); a fan-out of sessions multiplies spend.+- **Configuration stays CLI-managed.** Never read or hand-edit `.project_config.toml` or `agent_state/.vault.toml` — models and secrets go through `penguin config` (see the penguin-cli skill).diff --git a/packages/skills/skills/penguin-orchestration/icon.svg b/packages/skills/skills/penguin-orchestration/icon.svgnew file mode 100644index 00000000..7840ba49--- /dev/null+++ b/packages/skills/skills/penguin-orchestration/icon.svg@@ -0,0 +1,8 @@+<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="1.7" stroke-linecap="round" stroke-linejoin="round">+ <circle cx="12" cy="5.2" r="2.5" />+ <circle cx="5.5" cy="18.3" r="2.2" />+ <circle cx="18.5" cy="18.3" r="2.2" />+ <path d="M10.9 7.4 6.6 16.3" />+ <path d="M13.1 7.4 17.4 16.3" />+ <path d="M8.4 18.3h7.2" />+</svg>diff --git a/packages/skills/src/index.ts b/packages/skills/src/index.tsindex 06c390b5..12faabd0 100644--- a/packages/skills/src/index.ts+++ b/packages/skills/src/index.ts@@ -212,6 +212,7 @@ export const SKILL_GROUPS: SkillGroupInfo[] = [skills: ["penguin-sdk","penguin-cli",+ "penguin-orchestration","agenthub-models","vllm","ollama",diff --git a/packages/skills/test/skills.test.ts b/packages/skills/test/skills.test.tsindex 39af534a..7d58e1f2 100644--- a/packages/skills/test/skills.test.ts+++ b/packages/skills/test/skills.test.ts@@ -152,6 +152,7 @@ describe("loadSkillGroups / groupSkills", () => {expect(groups[2]!.skills.map((s) => s.name)).toEqual(["penguin-sdk","penguin-cli",+ "penguin-orchestration","agenthub-models","vllm","ollama",@@ -221,6 +222,7 @@ describe("loadSkillGroups / groupSkills", () => {skills: ["penguin-sdk","penguin-cli",+ "penguin-orchestration","agenthub-models","vllm","ollama",