Back to skill
Skillv1.0.0
ClawScan security
LYGO Champion: Δ9RA (RA) — The Wolf · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignFeb 11, 2026, 9:28 AM
- Verdict
- Benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill is internally consistent with its stated purpose (a persona advisor for signal-hunting); it requests no credentials, has no install steps, and the included scripts only read local pack files and print/validate the pack hash.
- Guidance
- This skill appears coherent and low-risk: it only bundles persona text and two small local scripts that validate and print the included pack hash. Before installing or running: (1) manually inspect the referenced files (canon.json, persona_pack.md) if you care about provenance; (2) be cautious about following any external verifier link (https://clawhub.ai/DeepSeekOracle/lygo-mint-verifier) — verify the site independently before posting anchors or running tools from it; (3) do not run the skill with elevated permissions or point it at system directories; and (4) remember the persona advocates 'censorship-breaking' investigative thinking but the behavior contract forbids illegal or invasive actions — ensure any investigations you perform remain lawful and ethical.
Review Dimensions
- Purpose & Capability
- okName, description, and included files are coherent: persona text, a local canon.json with a SHA-256 hash, and small helper scripts that read those local files. No unrelated credentials, binaries, or system access are requested.
- Instruction Scope
- okSKILL.md limits behavior to advisory OSINT/threat-modeling, insists on legality, and instructs showing the local LYGO-MINT hash and using an external verifier. There are no instructions to read arbitrary system files, exfiltrate secrets, or call unknown endpoints from the runtime instructions.
- Install Mechanism
- okNo install spec (instruction-only) — low-risk. The skill includes two small scripts that operate on the skill's own files; they do not download or execute external code.
- Credentials
- okThe skill requires no environment variables, credentials, or config paths. The local scripts only read files bundled in the skill (SKILL.md, references/*).
- Persistence & Privilege
- okalways:false (no forced presence). disable-model-invocation:false is the platform default and not itself concerning. The skill does not request system-wide config changes or other skills' credentials.