task-lifecycle
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill orchestrates a code development workflow that ingests data from external sources such as issue reports and pull requests, creating a potential surface for indirect instructions to influence agent behavior.
- Ingestion points: SKILL.md specifies that the agent should process user-supplied implementation tasks and existing PRs/branches.
- Boundary markers: The workflow implements strong isolation between roles, ensuring that reviewers and verifiers are fresh agents without access to the builder's context or reasoning.
- Capability inventory: The builder agent has the ability to write code to branches, while the verifier agent boots applications and interacts with non-production databases (SKILL.md, references/verify-round.md).
- Sanitization: The instructions explicitly forbid capturing real person's data and require human oversight for final merges and deployments.
Audit Metadata