loop-it

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

• Third-party content exposure detected (high risk: 0.85). Runtime path: the skill fetches GitHub issues via gh issue list ... --json ... --body, then reads/parses the outsider-authored issue body text to build the dependency graph and decide processing order; this free-form outsider text is ingested into the agent’s LLM context for parsing/logic.

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)