OpenClaw Task Experience Summaries
PassAudited by ClawScan on May 1, 2026.
Overview
This is an instruction-only troubleshooting reference; its command and credential examples are relevant but should be used deliberately.
This skill appears safe to install as a reference document. Before following its examples, verify package names, be cautious with force or publish commands, and do not save API keys or other secrets in the skill file.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
If a user runs these commands, they may install global packages, overwrite existing files during a forced reinstall, or publish a skill.
The skill documents commands that can change the local environment or publish content, but they are presented as troubleshooting and CLI usage examples rather than automatic actions.
npm install -g clawhub ... npm i -g clawhub --force ... clawhub publish "skill-directory"
Review each command before running it, avoid force flags unless needed, and confirm what will be published before using publish commands.
Users may handle real API keys or account tokens while following the guidance.
The skill discusses service tokens and API-key environment variables. This is expected for troubleshooting integrations, and the artifact does not show credential collection, logging, or transmission.
Token: `clh_<token>` ... set TAVILY_API_KEY=your-api-key-here ... `OPENAI_API_KEY`
Use least-privilege tokens, avoid pasting secrets into shared files or chat, and rotate any key that may have been exposed.
Incorrect or unreviewed notes could influence future troubleshooting decisions.
The skill encourages persistent updates to its own guidance. That is aligned with its experience-summary purpose, but future agents may rely on whatever gets written there.
Also use when documenting new experiences for future reference ... Update relevant section in this SKILL.md
Review additions before saving them, avoid storing secrets in the skill file, and clearly separate verified fixes from experimental observations.