start-issue
Pass
Audited by Gen Agent Trust Hub on Aug 11, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by sourcing operational logic and patterns from repository-controlled configuration files.
- Ingestion points: The agent reads docs/agents/issue-tracker.md and docs/agents/vcs.md from the local repository to determine branching patterns and status update methods.
- Boundary markers: No boundary markers or safety instructions are present; the agent is not instructed to treat the repository content as potentially untrusted data.
- Capability inventory: The agent has the capability to modify the local file system via git commands (checkout, pull, branch) and perform network requests to interact with issue tracking systems.
- Sanitization: The skill lacks any mechanism to sanitize or validate the regex patterns or method descriptions retrieved from the repository files before using them in command construction or tool calls.
Audit Metadata