Create, Evaluate, and Connect AI Skills
Python ai-agentsai-skillsartificial-intelligenceknowledge-graphlarge-language-modelsnatural-language-processingopenkgskillnetskills
update README.md
Jiang Chencfdaaf329d03
diff --git a/README.md b/README.mdindex b07d92e..79c6b19 100644--- a/README.md+++ b/README.md@@ -32,13 +32,13 @@ Search 500+ community skills · One-line install · Auto-create from repos / doc## ✨ Key Features-| Feature | Description |-| :--- | :--- |-| 🔍 **Search** | Find skills via keyword match or AI semantic search across 500+ curated skills |-| 📦 **One‑Line Install** | `skillnet download <url>` — grab any skill from GitHub in seconds |-| ✨ **Auto‑Create** | Convert GitHub repos, PDFs/PPTs/Word docs, conversation logs, or text prompts into structured skill packages using LLMs |-| 📊 **5‑D Evaluation** | Score skills on **Safety · Completeness · Executability · Maintainability · Cost‑Awareness** |-| 🕸️ **Skill Graph** | Auto-discover `similar_to` · `belong_to` · `compose_with` · `depend_on` links between skills |+| Feature | Description |+| :-------------------------------------- | :---------------------------------------------------------------------------------------------------------------------- |+| 🔍 **Search** | Find skills via keyword match or AI semantic search across 500+ curated skills |+| 📦 **One‑Line Install** | `skillnet download <url>` — grab any skill from GitHub in seconds |+| ✨ **Auto‑Create** | Convert GitHub repos, PDFs/PPTs/Word docs, conversation logs, or text prompts into structured skill packages using LLMs |+| 📊 **5‑D Evaluation** | Score skills on **Safety · Completeness · Executability · Maintainability · Cost‑Awareness** |+| 🕸️ **Skill Graph** | Auto-discover `similar_to` · `belong_to` · `compose_with` · `depend_on` links between skills |---@@ -362,19 +362,89 @@ Install the skillnet skill from ClawHub.</details>+<details>+<summary><b>⚙️ Configuration (Silent Pre-Configuration)</b></summary>++Three parameters control how SkillNet runs inside OpenClaw. If pre-configured in `openclaw.json`, the agent uses them silently — no prompts, no interruptions. If not configured, the agent only asks when a command actually needs the value, injects it for that single call, and never pollutes the global environment.++| Parameter | Required For | Default |+| :------------- | :-------------------------------- | :-------------------------- |+| `apiKey` | `create` · `evaluate` · `analyze` | — |+| `BASE_URL` | Custom / local LLM endpoint | `https://api.openai.com/v1` |+| `GITHUB_TOKEN` | Private repos / rate-limit bypass | — |++> `search` and `download` (public repos) work without any credentials.++**Recommended: pre-configure in `openclaw.json`** (save & restart gateway):++```json+{+ "skills": {+ "entries": {+ "skillnet": {+ "enabled": true,+ "apiKey": "sk-REPLACE_ME",+ "env": {+ "BASE_URL": "https://api.openai.com/v1",+ "GITHUB_TOKEN": "ghp_REPLACE_ME"+ }+ }+ }+ }+}+```++**Common scenarios:**++- **OpenAI only (minimal):** fill `apiKey` only; remove or keep `env` block with defaults.+- **Local inference (vLLM / LM Studio / Ollama):** set `apiKey` to any placeholder (e.g. `"sk-local"`), set `BASE_URL` to `"http://127.0.0.1:8000/v1"`.+- **Private repos / rate limits:** add a read-only `GITHUB_TOKEN` (recommended scope: `repo:read` or fine-grained read-only).++**Without pre-configuration (on-demand prompting):**++- First time the agent runs `create` / `evaluate` / `analyze` without a configured `apiKey` → it asks once in chat, you reply, and the key is used only for that call.+- Accessing a private GitHub repo or hitting rate limits → the agent asks for a read-only `GITHUB_TOKEN` only when needed.++</details>+<details><summary><b>🧪 Quick Verification</b></summary>In your OpenClaw chat, try:+**No credentials needed:**+```Search SkillNet for a "docker" skill and summarize the top result.```+**Requires API key (verifies silent injection):**+```Create a skill from this GitHub repo: https://github.com/owner/repo (then evaluate it).```+> If pre-configured in `openclaw.json`, the process runs silently end-to-end. If not, the agent asks for an API key once.++</details>++<details>+<summary><b>🔒 Security & Privacy</b></summary>++- **API key** is only sent to the `BASE_URL` you configured (default: OpenAI). It is **never** sent to the SkillNet search service.+- **GitHub token** is only used to call `api.github.com` for reading repository contents.+- **Search & download** only send the query string to `https://api-skillnet.openkg.cn` — no local files or credentials are uploaded.++</details>++<details>+<summary><b>🛠️ Troubleshooting</b></summary>++| Symptom | Cause | Fix |+| :------------------------------------------------------------- | :---------------------------------------------------- | :----------------------------------------------------------------------------------- |+| `API_KEY environment variable is required for skill creation.` | `create`/`evaluate`/`analyze` triggered without a key | Add `apiKey` in `openclaw.json`, or provide it when the agent asks in chat |+| GitHub 403 / rate limit / private repo access denied | Missing or insufficient GitHub token | Add a read-only `GITHUB_TOKEN` in `openclaw.json` `env`, or provide it when prompted |+</details>> The skill source is also available at [`skills/skillnet/`](skills/skillnet/) for reference.@@ -395,6 +465,8 @@ You can also [open an Issue](https://github.com/zjunlp/SkillNet/issues) to repor----## License+## LicenseThis project is licensed under the [MIT License](LICENSE).++写清楚批量导入功能