qa-test-healer
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell commands to run tests in headed and debug modes as part of its core healing workflow in SKILL.md.
- [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection due to its ingestion of untrusted data from the browser and logs.
- Ingestion points: The agent ingests data from user-provided test paths, error logs, CI failure logs, browser accessibility snapshots, and network request logs as described in references/diagnosis-patterns.md.
- Boundary markers: There are no explicit delimiters or instructions provided to the agent to prevent it from following instructions that might be embedded within the DOM content or logs being analyzed.
- Capability inventory: The skill has the capability to write to the file system (updating Page Object Models and test scripts) and execute shell commands (running the test suite to verify fixes) as outlined in SKILL.md and references/fix-strategies.md.
- Sanitization: The skill lacks evidence of sanitization or validation for the content retrieved from external browser states or log files before processing it for diagnostic decisions.
Audit Metadata