write-agent-context-docs
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill promotes a 'Living Document' workflow where agents are instructed to synthesize information from external sources, such as documentation fetched from the web, and write this into persistent project files like
AGENTS.md. This creates a surface for indirect prompt injection, as malicious instructions embedded in third-party content could be incorporated into the project's permanent guidance and influence future sessions. - Ingestion points: External documentation fetched during web exploration or third-party tool analysis (mentioned in the Quality Checklist).
- Boundary markers: The skill does not provide specific instructions for using delimiters or 'ignore' instructions when summarizing external content.
- Capability inventory: The agent is expected to write and update markdown files on the local filesystem to maintain project context.
- Sanitization: No specific sanitization or validation of external content is mandated before synthesis into the context files.
- [NO_CODE]: The skill consists entirely of markdown instructions and documentation templates. It does not include any executable scripts, binaries, or automated code snippets.
Audit Metadata