update-docs
Pass
Audited by Gen Agent Trust Hub on Jun 30, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by instructing the agent to record conversation context and user-provided information into persistent instruction files (CLAUDE.md) and project memory files.
- Ingestion points: The skill ingests conversation context and user-supplied details regarding code fixes, features, or incidents to update documentation (SKILL.md).
- Boundary markers: There are no instructions to use boundary markers or "ignore embedded instructions" warnings when writing this data into the documentation files.
- Capability inventory: The skill employs file read and write operations targeting multiple CLAUDE.md files and the ~/.claude project memory directory (SKILL.md).
- Sanitization: There is no mention of sanitizing, escaping, or validating the input data before it is permanently stored in the agent's long-term instructions.
Audit Metadata