shipyard-executing-plans
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes implementation plans that serve as an ingestion point for external data into the agent's task-execution context.
- Ingestion points: Implementation plan files are loaded and reviewed in 'Step 1: Load and Review Plan' within
SKILL.md. - Boundary markers: The instructions lack explicit delimiters or structural guards when passing plan content to builder agents, although they require a 'critical review' by the agent before proceeding.
- Capability inventory: The skill uses
TaskCreateandTeamCreateto manage execution, and dispatches various sub-agents (builder,shipyard:auditor,shipyard:simplifier) with capabilities to modify code, run tests, and commit changes. - Sanitization: No explicit sanitization or instruction filtering for the plan content is described before it is used to drive agent actions.
Audit Metadata