dbengoa42-star/edgeclaw/tool-governance
Claude Code and agent skill by dbengoa42-star/edgeclaw
- community
Checking that the SKILL.md file is still in the repository.
Guidelines for safe tool usage under ClawXGovernor. Follow these rules to avoid dangerous operations and work efficiently with tool risk controls.
Install
npx vskill@latest install dbengoa42-star/edgeclaw/tool-governanceSKILL.md
The same SKILL.md was published earlier as dbengoa42-star/edgeclaw/tool-governance.
Tool Usage Governance
The ClawXGovernor plugin monitors all tool calls for safety. Follow these guidelines:
Risk Levels
| Level | Tools | Default Action | |
|---|---|---|---|
| read | find, grep, read, glob, search | Allow | |
| workspace_write | edit, write, apply_patch | Allow | |
| exec | exec, code_execution, shell | Requires approval | |
| network | web_search, fetch | Allow | |
| subagent_control | sessions_spawn, sessions_send | Requires approval |
Before Executing Shell Commands
- Consider whether the command could be destructive
- Never run commands that delete system files, format disks, or create fork bombs
- If unsure about a command's safety, use
clawxgovernor__tool_risk_classifyto check first
Loop Prevention
The governor blocks tool calls that repeat more than 3 times consecutively with identical parameters. If you get blocked for looping:
- Change your approach — try a different tool or different parameters
- If you genuinely need to repeat a call, explain why to the user first
Long Tool Results
When a tool returns more than 4000 characters of output:
- The governor automatically creates a summary in
~/.openclaw/cc-tool-governor/summaries/ - Reference the summary instead of repeating the full output
- If you need the full output, tell the user where the complete data is stored
Audit Trail
Use clawxgovernor__tool_audit_query to review recent tool call history:
- Check which tools were called and their risk classifications
- Review any blocked or approval-required calls
- Useful for debugging when tools seem to behave unexpectedly
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingextensions/clawxgovernor/skills/tool-governance/SKILL.md
Popularity
- 7-day trend score
- 5
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/dbengoa42-star/edgeclaw/tool-governance)