find-features
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to scan and interpret the entire repository codebase to identify feature areas, which introduces a potential attack surface for indirect prompt injection if the source code contains malicious instructions.\n
- Ingestion points: Phase 1 reads project-level configuration files (
AGENTS.md), and Phase 4 scans the full codebase for implementation details.\n - Boundary markers: The skill does not explicitly instruct the agent to use delimiters or security-aware parsing when ingesting external codebase strings.\n
- Capability inventory: The skill possesses the capability to write and modify multiple markdown files across the repository, including an index file (
FEATURES.md).\n - Sanitization: There are no specified procedures for sanitizing or validating extracted code strings before they are interpolated into the documentation templates.\n- [COMMAND_EXECUTION]: The skill uses a shell command for local file discovery.\n
- Evidence: In Phase 1, the skill uses the
findcommand (find . -maxdepth 3 -name AGENTS_FEATURES.md) as a fallback to locate the documentation contract. This execution is limited in scope and used for legitimate environment discovery.
Audit Metadata