Skillv1.0.2

ClawScan security

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

BenignMar 20, 2026, 10:53 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill is internally consistent with its stated purpose: small local responses, a tiny local script, and a static quotes file — no unexpected credentials, network endpoints, or installs.
Guidance: This skill is simple and appears low-risk: it runs a tiny local Python script that prints "Hello, World!" with a timestamp and returns a line from a bundled quotes file. Before installing: (1) confirm you are comfortable allowing the agent to run small bundled scripts (it will call the local `python` binary — ensure your environment provides it); (2) review the two included files (they are short and human-readable); (3) note there is no network access, credentials, or external downloads in the package. If you do not want the agent to call skills autonomously, consider disabling autonomous invocation for this skill in your agent settings (the skill itself uses default invocation settings). The package has no homepage and an owner ID only — that is not a security problem by itself but is a minor trust signal to consider if provenance matters to you.

Purpose & Capability: okName/description promise simple greeting behavior. Included files (hello script and a quotes file) match that purpose and are proportionate.
Instruction Scope: noteSKILL.md instructs running a local Python script and selecting lines from a bundled text file — both are within the skill's scope. Minor mismatch: SKILL.md calls `python scripts/hello-world.py` but the skill metadata did not declare any required binaries; this merely means the runtime must have a Python interpreter available. The script itself is short and only prints a timestamped greeting; references/praise.md is a static list of motivational quotes.
Install Mechanism: okNo install spec (instruction-only with two small bundled files). Nothing is downloaded or written to disk beyond the included files.
Credentials: okNo environment variables, credentials, or config paths are requested. The skill does not attempt to access unrelated secrets or system configs.
Persistence & Privilege: okalways:false (default) and agent invocation is allowed (normal). The skill does not request elevated or persistent platform privileges.