claw-helper
v1.0.7OpenClaw 助手。解答 OpenClaw 相关问题，使用 CLI 修复问题。当用户询问 OpenClaw 使用、配置、故障排除，或需要帮助使用 OpenClaw CLI 命令时使用此技能。
⭐ 0· 364·3 current·3 all-time
by要啥自行车@thincher
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name and description match the instructions: the SKILL.md focuses on diagnosing and fixing OpenClaw via the openclaw CLI and consulting/updating local documentation. Nothing in the skill asks for unrelated binaries, cloud creds, or external services.
Instruction Scope
The instructions explicitly tell the agent to run openclaw CLI commands (e.g., openclaw doctor, openclaw config set) and to read/priority-check documentation at a docsPath injected via the SysPrompt. It also instructs updating local docs/index and writing to {{openclaw-help路径}}/reference/experience.md. This is coherent with a troubleshooting helper, but it means the agent will be instructed to perform config changes and write files — actions that can be destructive if run without user approval.
Install Mechanism
There is no install spec and no code files — this is an instruction-only skill, so nothing will be downloaded or written to disk by the skill itself.
Credentials
The skill declares no required environment variables, credentials, or config paths. It does reference templated values (docsPath, openclaw-help path) expected to be provided by the agent SysPrompt; those should be verified but are reasonable for its purpose.
Persistence & Privilege
always:false and no install/persistent hooks. The skill asks the agent to modify configs via the openclaw CLI and to update local docs, but it does not request system-wide privileges or modify other skills' configs.
Assessment
This skill appears to do what it says, but it instructs the agent to run OpenClaw CLI commands that can change configuration and to write/update local documentation files. Before installing or invoking it: (1) confirm what values are provided for docsPath and the openclaw-help path so the agent writes to expected locations; (2) require explicit user approval before the agent runs commands that change configuration or restart services; (3) ensure backups or a staging environment exist if you plan to allow automatic fixes; and (4) if you do not want the agent to modify files automatically, restrict it to read-only troubleshooting guidance.Like a lobster shell, security has layers — review code before you run it.
latestvk97bqybkjdc3vasab43vd1t03582jyz5
License
MIT-0
Free to use, modify, and redistribute. No attribution required.