cw-dev-edit

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's primary function is to analyze user-provided text drafts and provide feedback on structure, argument, and evidence. No malicious patterns or security risks were identified.
  • [DATA_EXFILTRATION]: The skill references a local file at ../../references/context-contract.md to load project-specific context. This is a standard instructional pattern for obtaining necessary background information for the editing task and does not involve sensitive system directories, credentials, or network transmission.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied drafts, which constitutes a potential attack surface. However, the skill does not define any capabilities or tools (such as file writing, shell execution, or network requests) that could be exploited through this vector.
  • Ingestion points: User drafts processed by the agent instructions in SKILL.md.
  • Boundary markers: None explicitly defined for the draft content.
  • Capability inventory: No tools are utilized; output is restricted to natural language feedback reports.
  • Sanitization: Not present, as the output is descriptive feedback for a human writer.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 06:43 PM
Security Audit — agent-trust-hub — cw-dev-edit