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
Risk Level
SAFE
Analyzed
Aug 12, 2026, 02:24 PM
Security Audit — agent-trust-hub — c4-context