implement-issues
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from GitHub issues and 'agent briefs' to guide its implementation logic, which could be manipulated to influence agent behavior.
- Ingestion points: Reads externally-provided issue descriptions and
AGENT-BRIEF.mdfiles from the repository. - Capability inventory: Writes to the local filesystem (git worktree), executes shell commands via
git,gh, andjq, and runs project-specific test and lint suites. - Boundary markers: The instructions mandate following the brief as a contract but do not provide explicit delimiters or instructions to ignore potential injections within that content.
- Sanitization: There is no explicit requirement for the agent to sanitize or validate the content of the brief before implementation.
- [COMMAND_EXECUTION]: The skill executes local commands such as
gh,git, andjq, and triggers project-level verification tools (tests, linters). While these are integral to the skill's purpose, they run within the environment and interact with external repository state.
Audit Metadata