Skillv0.1.0

ClawScan security

Code Architect · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignApr 6, 2026, 7:20 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: This is an instruction-only skill that plausibly implements its stated purpose (produce minimal implementation plans from the local codebase); it makes no installs and requests no credentials, so its footprint is coherent with its description.
Guidance: This skill is instruction-only and coherent with its description: it will analyze your workspace and propose minimal implementation plans. Before installing or invoking it, consider: (1) the agent will read repository files — remove or avoid placing secrets in the workspace you don't want included in analysis; (2) review any plan output before applying changes to code or CI; (3) if you want tighter control, run the agent in a sandbox or explicitly restrict which files it can access or ask the skill to list which files it needs to read before proceeding. If the skill later requests network access, credentials, or an install script, treat that as a new risk and reevaluate.

Purpose & Capability: okThe name and description match the SKILL.md: the skill's purpose is to produce minimal implementation plans that fit an existing codebase. It is instruction-only and does not request unrelated credentials, binaries, or config paths, which is proportionate to that purpose.
Instruction Scope: noteThe instructions direct the agent to use the 'existing local codebase' and workspace context to create plans. They do not instruct network exfiltration or reading unrelated system secrets, but they imply the agent will read repository files and context. Users should be aware that any sensitive material in the workspace could be included in analysis outputs.
Install Mechanism: okNo install spec and no code files are provided (instruction-only), which is the lowest-risk installation model — nothing will be written to disk by an installer.
Credentials: okThe skill requests no environment variables, credentials, or config paths. That aligns with a purely advisory, design-focused skill that analyzes local code.
Persistence & Privilege: okThe skill does not request always:true, does not modify other skills or system-wide settings, and uses the default autonomous-invocation setting (normal). It does not require elevated persistence or privileges.