bro-give-me-plan
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided code and context to generate implementation plans, creating a surface for indirect prompt injection where malicious instructions embedded in the analyzed code could influence the agent's behavior.
- Ingestion points: The skill reads code and environment context from the user session to gather information (referenced in SKILL.md and inline-plan.md).
- Boundary markers: The instructions do not specify the use of delimiters or 'ignore' commands for the untrusted data, although the agent is constrained by a rigid output template.
- Capability inventory: The agent has capabilities to read files, write markdown files to user-specified paths, and call specific tools like CreatePlan and AskQuestion.
- Sanitization: No sanitization or validation protocols for the ingested code context are defined in the instructions.
Audit Metadata