plan-audit
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (HIGH): The skill processes untrusted data from markdown plans and source code to perform complex logical audits.
- Ingestion points: Workflow steps 1 and 3 in
SKILL.mdidentify reading local files indocs/codex-plans/*.mdand inspecting repository state viagit log,git show, andrg. - Boundary markers: Absent. The instructions do not provide delimiters or safety constraints to protect the agent from instructions embedded in the external plan files or codebase.
- Capability inventory: Logic analysis and reporting findings. Malicious input in plans can cause the agent to provide false security assurances or intentionally overlook gaps in implementation (Silent Failure).
- Sanitization: Absent. No validation or filtering is performed on the plan text or code before it is processed by the agent's reasoning engine.
Recommendations
- AI detected serious security threats
Audit Metadata