workspace-docs
Pass
Audited by Gen Agent Trust Hub on Jun 27, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill instructs the agent to search for and read various documentation files (e.g.,
AGENTS.md,CLAUDE.md,.docs/instructions.md) which may contain untrusted content. - Ingestion points: The agent reads content from
.docs/instructions.md,.docs/instructions.append.md,AGENTS.md,CLAUDE.md,README.md,SPEC.md,SPECIFICATION.md, and files matchingdocs/*.mdusing thereadandfindtools. - Boundary markers: There are no instructions to wrap the read content in protective delimiters or to treat the instructions therein as untrusted.
- Capability inventory: The agent uses the gathered information to influence subsequent code creation, implementation tasks, and bug fixes.
- Sanitization: No sanitization or validation of the documentation content is performed before the agent acts upon it.
Audit Metadata