gate
Pass
Audited by Gen Agent Trust Hub on Aug 7, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches quality gate status and detailed issue reports (Bugs, Vulnerabilities, Hotspots, and Code Smells) from SonarCloud's official API endpoints.\n- [COMMAND_EXECUTION]: Uses
curlto query external APIs andgh run watchto track the status of GitHub Actions workflows. It also performs shell environment manipulation (unset GITHUB_TOKEN) to manage the authentication context for the GitHub CLI.\n- [DATA_EXFILTRATION]: Retrieves project-specific metadata and code quality metrics from SonarCloud. No local sensitive information or credentials are transmitted to external servers.\n- [PROMPT_INJECTION]: Represents an indirect prompt injection surface as it processes external data from SonarCloud APIs to inform code modifications.\n - Ingestion points: Data enters the agent context through JSON responses from SonarCloud API endpoints specified in SKILL.md.\n
- Boundary markers: No explicit delimiters or instructions to ignore embedded content are used when processing API results.\n
- Capability inventory: The skill can modify source code, execute
ghcommands, and monitor CI/CD pipelines as described in SKILL.md.\n - Sanitization: There is no explicit sanitization or validation of the text content returned by the SonarCloud API before it is processed by the agent.
Audit Metadata