matt-implement-spec
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from external specifications, issue tickets, and documentation, which could be used to deliver malicious instructions.
- Ingestion points: The agent is instructed to read the "spec", "tickets", "research notes", and "external documentation" to understand tasks (SKILL.md, Steps 1 and 2).
- Capability inventory: The skill performs file system modifications, executes git commands (branching, merging, worktrees), and orchestrates subagents to perform testing and code review (SKILL.md, Steps 4, 7, and 8).
- Boundary markers: There are no explicit instructions or delimiters defined to separate untrusted data from the agent's operational logic or to warn the agent about ignoring embedded instructions.
- Sanitization: No content validation or sanitization mechanisms are specified for the external data ingested from the project tracker or documentation.
Audit Metadata