snap-agents-md
Pass
Audited by Gen Agent Trust Hub on May 16, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted repository content to draft instructions.
- Ingestion points: Reads existing AGENTS.md, CLAUDE.md, manifests, workspace configs, and lockfiles in SKILL.md.
- Boundary markers: Lacks explicit boundary markers or instructions to disregard embedded commands in the source files.
- Capability inventory: Capable of writing AGENTS.md files and deleting existing instruction files (SKILL.md).
- Sanitization: No sanitization or content validation is performed on the ingested project data.
- [COMMAND_EXECUTION]: Uses local CLI utilities for project exploration and file discovery.
- Evidence: Uses fd, find, and rg tools to inventory repository files and resolve symlinks as specified in SKILL.md.
Audit Metadata