arinhub-verify-requirements-coverage
Pass
Audited by Gen Agent Trust Hub on Feb 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection through external data ingestion.
- Ingestion points: Reads pull request bodies via
gh pr viewand issue descriptions viagh issue view(Steps 3, 5). - Boundary markers: Absent. The procedure does not define delimiters or provide instructions for the agent to ignore or isolate embedded commands found within the issue or PR text.
- Capability inventory: The skill can execute
gitandghcommands, allowing it to read code, metadata, and repository state. - Sanitization: Absent. Data from these external sources is directly parsed for requirements (Step 6) and used to generate analysis reports.
- [COMMAND_EXECUTION]: The skill constructs shell commands using variables derived from user input and remote metadata.
- Variables like
$PR_NUMBER,$ISSUE_NUMBER, and${MERGE_BASE}are used in commands such asgh pr view $PR_NUMBERandgit diff "${MERGE_BASE}". If the agent fails to strictly extract numeric or valid git identifiers as instructed in Step 2, this could lead to command injection.
Audit Metadata