implement
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon untrusted data provided in the form of specifications or tickets.
- Ingestion points: External work descriptions, referred to as a "spec or set of tickets" in
SKILL.md, serve as the primary input for the skill's logic. - Boundary markers: Absent. The instructions do not define delimiters or warnings to prevent the agent from obeying instructions embedded within the user-provided specifications.
- Capability inventory: The skill is authorized to run typechecking tools, execute test suites, and perform git commits to the current repository branch (
SKILL.md). - Sanitization: Absent. There are no instructions to validate, filter, or sanitize the content of the specifications before the agent processes them.
Audit Metadata