Create, Evaluate, and Connect AI Skills
Python ai-agentsai-skillsartificial-intelligenceknowledge-graphlarge-language-modelsnatural-language-processingopenkgskillnetskills
update README.md
Jiang Chen803fc933bf3e
diff --git a/README.md b/README.mdindex d02fcae..2f002fb 100644--- a/README.md+++ b/README.md@@ -44,15 +44,13 @@ SkillNet is an open infrastructure for creating, evaluating, and organizing AI s- GitHub repositories- Office documents (PDF, PPT, Word)- Direct text prompts-- **π Evaluation**: Evaluate and score skills for quality assurance (Safety, Completeness, Excutability, Maintainability, Cost-Awareness).+- **π Evaluation**: Evaluate and score skills for quality assurance (Safety, Completeness, Executability, Maintainability, Cost-Awareness).- **πΈοΈ Relationship Analysis**: Automatically map the connections between skills in your local library, identifying structural relationships between skills (similar_to, belong_to, compose_with, depend_on).## π¬ Quick Start Demohttps://github.com/user-attachments/assets/9f9d35b0-36fd-4d7d-a072-39afa380b241--# π API AccessSkillNet provides a public API to search skills. Support both keywords match and semantic search.@@ -263,13 +261,13 @@ The CLI is powered by `Typer` and `Rich` for a beautiful terminal experience.#### Common Commands-| Command | Action | Example |-| :------------- | :------------- | :------------------------------------------- |-| **`search`** | Search skills | `skillnet search "data viz" --mode vector` |-| **`download`** | Install skill | `skillnet download <github_url> -d ./skills` |-| **`create`** | Create skill | `skillnet create log.txt --model gpt-4o` |-| **`evaluate`** | Evaluate skill quality| `skillnet evaluate ./my_tool` |-| **`analyze`** | Analyze skill relations | `skillnet analyze ./my_agent_skills` |+| Command | Action | Example |+| :------------- | :---------------------- | :------------------------------------------- |+| **`search`** | Search skills | `skillnet search "data viz" --mode vector` |+| **`download`** | Install skill | `skillnet download <github_url> -d ./skills` |+| **`create`** | Create skill | `skillnet create log.txt --model gpt-4o` |+| **`evaluate`** | Evaluate skill quality | `skillnet evaluate ./my_tool` |+| **`analyze`** | Analyze skill relations | `skillnet analyze ./my_agent_skills` |> **Tip:** Use `skillnet [command] --help` to see all available options (e.g., thresholds, sorting).@@ -437,6 +435,7 @@ To demonstrate the power of SkillNet, we provide a complete example of a Scientiπ **[Try the Interactive Demo (Website: Scenarios->Science)](http://skillnet.openkg.cn/)**### 𧬠Workflow Lifecycle+- Task Definition: User provides a goal: "Analyze scRNA-seq data to find cancer targets."- AI Planning: The Agent decomposes the mission into logical steps (Data $\rightarrow$ Mechanism $\rightarrow$ Validation $\rightarrow$ Report).- Skill Discovery: The Agent uses client.search() to find specialized skills like cellxgene-census and kegg-database.@@ -444,6 +443,7 @@ To demonstrate the power of SkillNet, we provide a complete example of a Scienti- Execution: The skills are executed sequentially to generate a final discovery report.### π Try it yourself+You can explore the full implementation, in our interactive Jupyter Notebook:π **[View Scientific Discovery Demo (Notebook)](https://github.com/zjunlp/SkillNet/blob/main/examples/scientific_workflow_demo.ipynb)**@@ -452,63 +452,53 @@ You can explore the full implementation, in our interactive Jupyter Notebook:## π€ OpenClaw Integration-SkillNet integrates with [OpenClaw](https://github.com/openclaw/openclaw) (an open-source personal AI agent framework) as a built-in, lazy-loaded skill. **You don't need OpenClaw's source code here** β just a working OpenClaw installation.--Once set up, the agent automatically:+SkillNet integrates with [OpenClaw](https://github.com/openclaw/openclaw) (open-source personal AI agent framework) as a built-in, lazy-loaded skill. You do **not** need OpenClaw's source code β only a working OpenClaw install. Once loaded, the agent automatically:- **Searches & applies** existing skills before tackling complex or unfamiliar tasks-- **Creates skills** from GitHub repos, documents (PDF/PPT/Word), or completed work-- **Evaluates & analyzes** the local skill library for quality scores and inter-skill relationships+- **Creates skills** from GitHub repos, documents (PDF/PPT/Word), or completed work via `skillnet create`+- **Evaluates & analyzes** your local skill library for quality scores and inter-skill relationships-<div align="center"> <img src="images/openclaw_skillnet.gif" width="100%" alt="OpenClaw + SkillNet Demo"> </div>+This forms a closed loop: community skills guide execution β successful outcomes become new skills β periodic analysis keeps the library clean.-### Quick Setup+<div align="center"> <img src="images/openclaw_skillnet.gif" width="100%" alt="OpenClaw + SkillNet Demo"> </div>-**Prerequisites**: OpenClaw is installed and running on your machine.+### π Prerequisites-**Step 1: Copy the skill to OpenClaw**+- [OpenClaw](https://github.com/openclaw/openclaw) installed with a configured workspace (default: `~/.openclaw/workspace`)+- No OpenClaw repo checkout is required-macOS/Linux:+### π₯ Install via ClawHub (Recommended)```bash-mkdir -p ~/.openclaw/skills-cp -R skills/skillnet ~/.openclaw/skills/skillnet-```+# 1. Install ClawHub CLI (once)+npm i -g clawhub-Windows PowerShell:+# 2. Install the skill into your OpenClaw workspace (pick one)+cd ~/.openclaw/workspace && clawhub install skillnet+# β or from anywhere β+clawhub install skillnet --workdir ~/.openclaw/workspace-```powershell-New-Item -Path "$env:USERPROFILE\.openclaw\skills" -ItemType Directory -Force-Copy-Item -Recurse "skills\skillnet" "$env:USERPROFILE\.openclaw\skills\skillnet"+# 3. Restart OpenClaw so it discovers the new skill+openclaw gateway restart # or simply start a new chat session```-**Step 2: Restart OpenClaw**+The skill lands at `<workspace>/skills/skillnet` β workspace skills take precedence over global and bundled ones.-```bash-openclaw gateway restart-```+### π§ͺ Quick Smoke Test-OpenClaw will detect the skill and prompt you to install the `skillnet` CLI automatically (choose from script/uv/pipx).+In your OpenClaw chat, try:-**Step 3: Trigger the skill in OpenClaw**--In OpenClaw chat, try:--> "Search SkillNet for a 'docker' skill and show the top result."--OpenClaw will detect that the `skillnet` CLI is missing and prompt you to install it (choose from script/uv/pipx). After installation completes, the command will execute automatically.--**Optional: Environment variables for advanced features**--Only needed if you use `create`, `evaluate`, or `analyze` commands:--- `API_KEY` β OpenAI-compatible API key (required for create/evaluate/analyze)-- `BASE_URL` β Custom LLM endpoint (optional, defaults to OpenAI)-- `GITHUB_TOKEN` β GitHub PAT for private repos or higher rate limits (optional)+```+Search SkillNet for a "docker" skill and summarize the top result.+```-**Note**: `search` and `download` are completely free β no keys required.+```+Create a skill from this GitHub repo: https://github.com/owner/repo (then evaluate it).+```-π The full integration skill package is available at [`skills/skillnet/`](skills/skillnet/).+> For SkillNet CLI installation and usage (`search` / `create` / `evaluate` / `analyze`), see the [Python Toolkit](#-python-toolkit-skillnet-ai) section above.+>+> π The source of this skill package is also available at [`skills/skillnet/`](skills/skillnet/) for reference, but the recommended install method is via **ClawHub** as shown above.---diff --git a/skills/skillnet/SKILL.md b/skills/skillnet/SKILL.mdindex 34997a3..7bfa603 100644--- a/skills/skillnet/SKILL.md+++ b/skills/skillnet/SKILL.md@@ -102,6 +102,8 @@ Download a skill and **immediately** load it into your current context:```bash# Download to local skill library+# <skill-url> must point to a specific skill subdirectory, e.g.:+# https://github.com/openclaw/openclaw/tree/main/skills/summarizeskillnet download "<skill-url>" -d ~/.openclaw/skills# Read SKILL.md β the skill's core instructions (always read this first)diff --git a/skills/skillnet/references/workflow-patterns.md b/skills/skillnet/references/workflow-patterns.mdindex d442932..ffcce2d 100644--- a/skills/skillnet/references/workflow-patterns.md+++ b/skills/skillnet/references/workflow-patterns.md@@ -122,6 +122,6 @@ Routine tasks, minor fixes, or straightforward work β do NOT create.| User provides a GitHub repo to learn | **create** (github) | `skillnet create --github <url>` || Finished a complex task with lessons | **create** (prompt) | `skillnet create --prompt "..."` || User shares a knowledge document | **create** (office) | `skillnet create --office <file>` |-| User provides execution logs or data | **create** (trajectory) | `skillnet create <trajectory-file>` |+| User provides execution logs or data | **create** (trajectory) | `skillnet create --trajectory <trajectory-file>` || Unsure about a skill's quality | **evaluate** | `skillnet evaluate <path-or-url>` || Too many skills, need organization | **analyze** | `skillnet analyze <dir>` |diff --git a/skills/skillnet/scripts/create_skill.py b/skills/skillnet/scripts/create_skill.pyindex c904c98..81ff535 100644--- a/skills/skillnet/scripts/create_skill.py+++ b/skills/skillnet/scripts/create_skill.py@@ -16,6 +16,8 @@ Usage:Requires: pip install skillnet-aiEnvironment variable: API_KEY+ Optional: --model to override default (gpt-4o)+ Optional: BASE_URL for OpenAI-compatible endpoints (default: https://api.openai.com/v1)"""import argparseimport osdiff --git a/skills/skillnet/scripts/search_and_download.py b/skills/skillnet/scripts/search_and_download.pyindex 6eab958..db7d1b5 100644--- a/skills/skillnet/scripts/search_and_download.py+++ b/skills/skillnet/scripts/search_and_download.py@@ -3,7 +3,7 @@Usage:python search_and_download.py "query" [--mode vector] [--threshold 0.7] \- [--download] [--target-dir ~/.openclaw/skills]+ [--download] [--target-dir ~/.openclaw/skills] [--no-fallback]Requires: pip install skillnet-ai"""@@ -23,10 +23,13 @@ def main():parser.add_argument("--download", action="store_true", help="Auto-download the top result")parser.add_argument("--target-dir", default=os.path.expanduser("~/.openclaw/skills"),help="Directory to install downloaded skills into")- parser.add_argument("--fallback", action="store_true", default=True,- help="If keyword returns 0 results, retry with vector mode")+ parser.add_argument("--no-fallback", action="store_true",+ help="Disable automatic retry with vector mode when keyword returns 0 results")args = parser.parse_args()+ # Determine effective fallback flag+ fallback = not args.no_fallback+try:from skillnet_ai import SkillNetClientexcept ImportError:@@ -47,7 +50,7 @@ def main():)# Fallback: keyword β vector- if not results and args.mode == "keyword" and args.fallback:+ if not results and args.mode == "keyword" and fallback:print(" No keyword results. Retrying with vector mode (threshold=0.65)...")results = client.search(q=args.query,