status
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from GitHub issue and pull request bodies to identify unfinished tasks and audit template compliance. An attacker could embed malicious instructions in these fields to influence the agent's behavior during the triage or recommendation phases.\n
- Ingestion points: PR and Issue bodies retrieved via
ghCLI commands inSKILL.md.\n - Boundary markers: Absent; the skill parses raw content to count unchecked task boxes and verify metadata.\n
- Capability inventory: The skill has the ability to modify repository state via
gh issue edit,gh issue close,gh pr edit, andgh pr close.\n - Sanitization: Absent; the skill relies on manual user confirmation (
AskUserQuestion) before any write operations are performed.\n- [COMMAND_EXECUTION]: The skill utilizes thegh(GitHub CLI) tool to perform repository management tasks. While the usage follows standard patterns, it grants the agent the capability to interact with and modify remote repository metadata and states.
Audit Metadata