c4-context
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security vulnerabilities were detected in the skill instructions or templates.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied system information to generate structured documentation. While it lacks explicit boundary markers for external data, the risk is negligible given its purpose as a documentation tool.
- Ingestion points: Processes system descriptions, feature lists, and persona details provided by the user during interaction (SKILL.md).
- Boundary markers: Absent; the skill relies on the agent's default handling of user input.
- Capability inventory: Generates markdown documentation and Mermaid context diagrams.
- Sanitization: None observed, which is typical for documentation-focused skills.
Audit Metadata