suggesting-next-steps
Pass
Audited by Gen Agent Trust Hub on Mar 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute various read-only CLI commands using
gitandghto gather context about the repository's state, including commit history, local stashes, and GitHub issue/PR details.\n- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it ingests and processes untrusted data from project documentation and issue trackers to generate task priorities.\n - Ingestion points: Processes content from files such as
STATUS.md,ROADMAP.md,TODO.md,README.md,package.json, and requirement documents indocs/, as well as GitHub issue and pull request descriptions.\n - Boundary markers: There are no explicit instructions to use delimiters or ignore embedded instructions within the parsed project files or GitHub content.\n
- Capability inventory: The skill restricts the agent to read-only capabilities via the
gitandghtools and file system access.\n - Sanitization: No sanitization or validation of the extracted external text is performed before it is used by the agent to determine the next steps.
Audit Metadata