plan-harder
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it is instructed to analyze codebase content which could contain adversarial instructions designed to influence the resulting implementation plan.
- Ingestion points: Untrusted data enters the agent context during Phase 0 (codebase investigation) and Phase 1 (request_user_input).
- Boundary markers: The skill does not define specific delimiters or instructions to ignore potential commands embedded within the files being researched.
- Capability inventory: The agent can perform file system reads for research, file system writes for saving plans, and invoke subagents for review, though it is forbidden from implementation.
- Sanitization: There are no mentions of content sanitization, escaping, or validation for the data ingested from the codebase or user queries.
Audit Metadata