Back to skill
Skillv1.0.0

ClawScan security

Changelog Release Manager · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignApr 29, 2026, 11:47 PM
Verdict
benign
Confidence
high
Model
gpt-5-mini
Summary
The skill is an instruction-only changelog generator whose described purpose matches its minimal requirements and footprint; it is internally coherent but the runtime instructions are very terse.
Guidance
This skill is coherent and low-risk as presented, but its runtime behavior is underspecified. Before installing or allowing it to run autonomously, confirm how it will obtain commit history (will it run local git commands, read a remote repo, or ask you to paste commits?) and require explicit permission for any file-system or network access. If you care about sensitive data, ensure the agent is prevented from sending repository contents to external services and that it only reads the repository you intend it to operate on.

Review Dimensions

Purpose & Capability
okName, description, and SKILL.md all describe generating changelogs from conventional commits. The skill requests no binaries, env vars, or install steps — which is consistent for an instruction-only helper that expects the agent to operate on available git history.
Instruction Scope
noteSKILL.md is brief and provides formatting guidance and commit-type mappings but does not specify HOW to obtain commit history (e.g., run `git` locally, read an API, or accept pasted commits). This makes runtime behavior ambiguous: a reasonable implementation would read local git history or ask the user for repo access, but the instructions give the agent broad discretion unless higher-level agent policies constrain it.
Install Mechanism
okNo install spec or code files are provided (instruction-only). That minimizes on-disk risk and is appropriate for a prose-guidance skill.
Credentials
okThe skill requests no environment variables, credentials, or config paths. That is proportionate to a changelog generator which does not inherently require secrets or external service access.
Persistence & Privilege
okThe skill is not always-enabled and does not request elevated persistence. Autonomous invocation is permitted by platform default but is not combined with other concerning privileges here.