Skillv1.0.0

ClawScan security

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

BenignApr 30, 2026, 7:25 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill is an instruction-only learning methodology that asks the agent to create study files and directories and otherwise contains no unrelated network, credential, or install behavior — its requirements align with its stated purpose.
Guidance: This is an instruction-only learning skill that will create a topic directory and several markdown files under workspace/study/{topic}/ if invoked. It does not request credentials or install code. Before installing, confirm you are comfortable with the agent writing files into your workspace path and review SKILL.md to ensure the file structure and templates match your expectations. If provenance matters, verify the author/source (README points to a GitHub repo) before trusting or sharing sensitive materials created by the skill.

Purpose & Capability: okName/description (learning methodology) match the runtime instructions and file templates. All requested actions (create topic folder, generate study documents) are coherent with a study-system skill.
Instruction Scope: okSKILL.md instructs the agent to guide users through a 9-step learning flow and create files under workspace/study/{topic}/. It does not instruct reading system config, accessing credentials, contacting external endpoints, or collecting unrelated data. File creation is explicit and justified by the skill's purpose.
Install Mechanism: okNo install spec or external downloads; the package is instruction-only. package.json and README exist for metadata but no code is executed or written by an installer.
Credentials: okNo environment variables, credentials, or config paths are requested. Recommended third-party tools are only suggestions and not required for operation.
Persistence & Privilege: okSkill does not request always:true or any elevated/omnipresent privilege. It will create and write files under its workspace path (documented) but does not modify other skills or system-wide agent settings.