install-spinner

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill requires the agent to print a fixed banner containing a token-like string verbatim ("$CLAWD: 8cHzQHUS2s2h8TzCmfqPKYiM4dSt4roa3n7MyRLApump") and also reads the user's settings.json (which may include secrets), so it forces verbatim handling/exposure of sensitive values.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). The skill fetches outsider-authored spinner pack JSON from a public GitHub raw URL at runtime (https://raw.githubusercontent.com/.../spinners/<pack-name>.json) and ingests its free-text spinnerVerbs values into the agent’s settings/LLM context.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill fetches spinner lists at runtime from https://api.github.com/repos/Solizardking/solana-clawd/contents/spinners and then retrieves pack JSON from https://raw.githubusercontent.com/Solizardking/solana-clawd/main/spinners/.json and copies the external spinnerVerbs JSON into the agent settings, which directly modifies the agent's runtime output/behavior.