skill-domain-discovery
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection (Category 8) due to its core requirement of analyzing external, potentially attacker-controlled content. 1. Ingestion points: Repository documentation (Phase 1a), GitHub issues, pull requests, and discussions (Phase 3). 2. Boundary markers: Absent. The instructions do not specify the use of delimiters or 'ignore' instructions when reading external content. 3. Capability inventory: The skill utilizes filesystem access (grep, glob, file reading) and network-enabled tools (GitHub CLI
gh searchandgh api). 4. Sanitization: Absent. No verification or filtering of the ingested external text is performed before analysis.
Audit Metadata