triage-issue
Pass
Audited by Gen Agent Trust Hub on Jun 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes shell commands
git logfor diagnostic investigation andgh issue createto automate the submission of triaged reports. These operations are consistent with the skill's primary objective of software maintenance. - [DATA_EXFILTRATION]: The skill extracts information about the codebase's root cause and fix plans, which are then transmitted to GitHub via the
ghtool. This is a legitimate use case for a triage automation skill. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted user input and internal codebase artifacts without explicit boundary markers or sanitization logic.
- Ingestion points: User-provided problem descriptions in Step 1 and codebase files/git history in Step 2 via the Explore subagent.
- Boundary markers: The instructions do not define clear delimiters or context-isolation markers for the ingested data.
- Capability inventory: The skill has the ability to execute subagents, query git history, and create external issues via the GitHub CLI.
- Sanitization: No input validation or output escaping is implemented before interpolating data into the issue template.
Audit Metadata