omakase-off
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input describing software features to be built and interpolates this into implementation plans and scenario tests.
- Ingestion points: User requests in Phase 0 and 1 (SKILL.md, references/detailed-workflow.md).
- Boundary markers: None identified; instructions do not specify delimiters or warnings to ignore embedded instructions in user requirements.
- Capability inventory: Invocation of subagents (parallel-agents), execution of git commands (git-worktrees), and execution of shell scripts for E2E testing (scenario-testing).
- Sanitization: None identified; user-provided content is passed to planning and implementation agents.
- [COMMAND_EXECUTION]: The skill utilizes shell commands via dependencies to manage the local file system and repository state.
- Evidence: usage of git worktree add and git worktree remove (references/detailed-workflow.md).
- [DYNAMIC_EXECUTION]: The skill generates and executes implementation plans and E2E scenario test scripts based on user-provided requirements.
- Evidence: The scenario-testing phase involves creating and running .scratch/ scripts (references/detailed-workflow.md).
Audit Metadata