dd-unblock-pr
Pass
Audited by Gen Agent Trust Hub on Jul 21, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes
gitcommands (git remote get-url,git symbolic-ref) to determine the repository identity and branch context. - [COMMAND_EXECUTION]: The skill utilizes the
pupCLI (a DataDog-provided tool) to fetch CI/CD events, test failure details, and code coverage data from DataDog's platform. - [COMMAND_EXECUTION]: The skill uses the GitHub CLI (
gh run rerun) to trigger retries of failed pipeline runs based on extracted run IDs. - [DATA_EXFILTRATION]: The skill transmits repository identifiers and branch names to DataDog's infrastructure via the
pupcommand-line tool. This is expected behavior given the skill's purpose for CI/CD health monitoring. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources, specifically CI error messages and test failure logs retrieved via
pup. As the skill lacks explicit boundary markers or sanitization for this content when generating the 'Triage Brief', there is a potential risk that malicious content within CI logs could influence the agent's summary or classification logic.
Audit Metadata