Back to skill
Skillv1.0.0
ClawScan security
usage-stats · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignApr 11, 2026, 5:56 PM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill reads local OpenClaw session logs and produces a local usage report; its code and instructions are consistent with that purpose and do not request external credentials or network access.
- Guidance
- This skill appears to do what it says: it parses your OpenClaw session logs and writes a usage report under ~/.qclaw/workspace/memory. Before running it, review the bundled script if you are concerned (it's pure Python, standard library only), and be aware the generated report may include excerpts or structured data derived from your conversations (potentially sensitive). If you don't want message contents included in reports, manually inspect/clean the session files or run the script in a sandboxed account. Because the skill has no network or credential access, it does not appear to exfiltrate data, but exercising caution with any tool that reads full conversation logs is recommended.
Review Dimensions
- Purpose & Capability
- okName/description claim to analyze OpenClaw session usage; the skill's SKILL.md and the included Python script both read ~/.qclaw/agents/main/sessions/*.jsonl and produce reports under ~/.qclaw/workspace/memory — this is coherent and proportional to the stated purpose.
- Instruction Scope
- noteInstructions explicitly tell the agent to run scripts/analyze_usage.py and read session JSONL files and existing history JSON; that stays inside the described purpose. Note: session files may contain sensitive user messages and the report can include message/error excerpts — users should expect local sensitive data will be parsed and summarized.
- Install Mechanism
- okNo install spec; script is bundled in the skill and uses only the Python standard library. No external downloads or package installs are requested.
- Credentials
- okThe skill requests no environment variables, no credentials, and accesses only paths under the user's ~/.qclaw directory as described. There are no unexpected or unrelated secrets requested.
- Persistence & Privilege
- okalways is false, user-invocable, and the skill writes its own output/history under ~/.qclaw/workspace/memory; it does not modify other skills or system-wide agent settings. Autonomous invocation is allowed by platform default but not unusually privileged here.