Skillv1.0.0

ClawScan security

Sovereign Project Guardian · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignFeb 23, 2026, 10:50 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's requested actions and data access are consistent with its stated purpose (auditing a project repository); it does not request excess credentials or install code, but it will read and analyze repository files (including secrets if present).
Guidance: This skill is coherent and does what it claims: it inspects a repository and reports problems. Before running it, consider: (1) run it against a copy or non-sensitive checkout if your repo contains secrets you don't want processed; the skill will scan all files and could surface secrets in its report; (2) expect it to recommend or run local audit tools that may require network access or installed tooling; (3) only allow the agent access to repos you trust, and monitor any outbound network activity from the agent if you are concerned about sensitive data leaving your environment. If you prefer, run the checks locally (using the examples in EXAMPLES.md) rather than granting the agent direct repository access.

Purpose & Capability: okName/description match the instructions: the SKILL.md describes repository discovery and a set of checks (security, quality, docs, CI/CD) and all required checks act on repository files and metadata. No unrelated binaries, credentials, or config paths are requested.
Instruction Scope: noteInstructions direct the agent to scan the entire repository (files, manifests, .git history, .gitignore) and to run or recommend dependency-audit tools (npm audit, pip-audit, govulncheck, cargo audit). This is appropriate for a project auditor, but it means the skill will process all file contents (including any secrets committed). It does not instruct the agent to read unrelated system files or transmit results to unknown external endpoints.
Install Mechanism: okInstruction-only skill with no install spec and no code files to write to disk. Lowest-risk installation footprint; no downloads, no brew/npm installs specified by the skill itself.
Credentials: okThe skill requires no environment variables or credentials. Any commands it recommends (e.g., npm audit) may require network access or installed tooling, but that is proportional to its auditing purpose.
Persistence & Privilege: okalways:false and no special config changes requested. The skill does not request permanent presence or modify other skills' configuration. Autonomous invocation is enabled by platform default but is not combined with other risky privileges.