implement-spec
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and process external specifications and tickets to drive the behavior of autonomous implementer, exploration, and merger subagents. This reliance on external content without robust structural boundaries introduces a risk where malicious instructions embedded in a spec could influence the subagents' actions.\n
- Ingestion points: The instructions in
SKILL.mddirect the agent to 'Read the spec and tickets' to understand a task graph.\n - Boundary markers: There are no explicit instructions or delimiters provided to separate the specification data from the agent's core operational instructions, nor any guidance to ignore embedded directives.\n
- Capability inventory: The orchestration involves significant capabilities including file system writes (implementation), branch and PR management, merging operations, and executing review tools like
/review-axes.\n - Sanitization: The skill lacks validation or sanitization steps for the external inputs before they are used to initialize subagent tasks.
Audit Metadata