implement-plan
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes architectural decision records (ADRs) and plan files which may contain untrusted data that could influence agent behavior.
- Ingestion points: The skill reads external plan files and ADR documents in Step 1 to understand the implementation scope.
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the plans are defined in the skill instructions.
- Capability inventory: The skill allows the agent to modify the file system (Step 5), execute verification checks and tests (Step 4), interact with a browser via
agent-browser(Step 4), and manage background processes/worktrees (Step 7). - Sanitization: There is no evidence of sanitization or schema validation performed on the plan files before the agent acts upon the instructions contained within them.
Audit Metadata