rhdh-plugin-bug-fix

Pass

Audited by Gen Agent Trust Hub on Aug 19, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted bug reports from external platforms such as Jira and GitHub. * Ingestion points: Workflow Step 1 in workflows/fix-bug.md resolves context from issue keys and URLs. * Boundary markers: The skill does not use specific delimiters to isolate ingested bug descriptions from its internal instructions. * Capability inventory: The agent has capabilities including file system access, shell command execution (yarn, playwright, ffmpeg), and network access via the GitHub CLI (gh). * Sanitization: No explicit sanitization or validation of the bug report content is mentioned before it is used to generate tests.
  • [DYNAMIC_EXECUTION]: The skill creates and runs temporary Playwright test scripts (_repro-<issue>.test.ts) based on inputs from bug reports, as described in workflows/fix-bug.md.
  • [COMMAND_EXECUTION]: The skill performs its operations by executing local CLI tools such as git, yarn, node, playwright, ffmpeg, and gh.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 19, 2026, 11:10 AM
Security Audit — agent-trust-hub — rhdh-plugin-bug-fix