Skillv1.0.0

ClawScan security

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

Scanner verdict

BenignApr 19, 2026, 1:57 AM

Verdict: Benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's declared purpose (generate SFT-level testcases for a target Skill) matches what its instructions and included helper script do; nothing requests unrelated credentials or performs network exfiltration.
Guidance: This skill appears coherent and implements what it claims: it inspects a target Skill's files and generates structured SFT testcases. Before running it, pick the target carefully — the tool will read all files under the target (including scripts and docs) and will extract any provided zip to a temp directory. Do not point it at directories containing secrets or credentials. Verify the generated JSON and the validate output locally before sharing it. Note the source is 'unknown' — if you require provenance, prefer skills from known authors or inspect generate.py yourself (it is included and readable) to confirm behavior.

Review Dimensions

Purpose & Capability: okName/description (generate structured SFT testcases) align with the included scripts and SKILL.md: the package provides a helper script (prepare/validate) and detailed runtime instructions for inspecting a target Skill and producing JSON test cases. Required capabilities (reading skill files, extracting zip) are appropriate for this purpose.
Instruction Scope: noteSKILL.md explicitly instructs the agent to read all files in the target Skill (SKILL.md, scripts/, references/) and to run the provided prepare/validate utilities. Reading all files is necessary for traceable test generation, but it is broad — the agent will inspect script files and any referenced docs, so users should avoid pointing it at targets containing secrets. The SKILL.md also hardcodes an example path (~/.workbuddy/skills/skill-test-generate/scripts/generate.py) which assumes a typical install location; this is a minor mismatch if the skill is present elsewhere.
Install Mechanism: okThere is no install spec (instruction-only with helper script included). No external downloads or archive extracts are performed by the skill installer. The included generate.py may extract a user-supplied zip to a temporary directory (expected behavior for preparing a zip-based skill).
Credentials: okThe skill does not request environment variables, credentials, or config paths. The helper script reads HOME and APPDATA to locate installed skills (benign behavior for 'installed:<name>' resolution). No secrets are required by design.
Persistence & Privilege: notealways:false and model invocation is allowed (default). The validate workflow is described to auto-generate a summary and add a 'verified' flag to the output JSON — this implies writing/updating the specified JSON file (expected). The script may extract zip archives to a temporary directory and may remove them on error; it does not request permanent system-wide privileges or modify other skills.