qa-bug-ticket-creator
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from test failure reports (JUnit XML, Playwright logs, pytest output) and manual defect descriptions. This data is then used to perform actions via GitHub MCP, such as searching for duplicates and creating new issues. A malicious actor could embed instructions within a test failure log (e.g., in a stack trace or failure message) to manipulate the agent's actions on GitHub.
- Ingestion points:
SKILL.md(Workflow steps 1 & 2) andreferences/bug-report-format.md(Mapping from Test Failure Formats section). - Boundary markers: Absent; there are no specific delimiters or instructions for the agent to ignore commands potentially embedded within the ingested test logs.
- Capability inventory: The skill leverages GitHub MCP for searching, linking, and creating issues with full body content, labels, and milestones.
- Sanitization: No mention of escaping, filtering, or sanitizing the content extracted from failure logs or stack traces before interpolation into the bug report format.
- [NO_CODE]: The provided files consist entirely of documentation, formatting guides, and Markdown templates. There are no executable scripts (e.g., Python, JavaScript, Shell) or compiled binaries, which significantly reduces the risk of direct malicious code execution or persistence.
Audit Metadata