plan-review
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external plan files to identify flaws and subsequently directs the agent to modify the plan and perform implementation based on those findings. This creates a vulnerability surface where malicious instructions could be embedded within a plan file to influence the agent's behavior during the review or implementation phase.
- Ingestion points: The sub-agent call in
SKILL.mdaccepts an absolute path to a plan file provided in the environment. - Boundary markers: There are no explicit delimiters or instructions defined in the skill to treat the plan file content as untrusted data.
- Capability inventory: The skill instructions lead to file modifications (fixing the plan) and code implementation (file writes and potentially command execution).
- Sanitization: The skill does not specify any sanitization, validation, or escaping of the content read from the plan file.
Audit Metadata