testing-reality-checker
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external test results, creating a surface where malicious instructions in the data could influence the agent's behavior.
- Ingestion points: The agent is instructed to read
public/qa-screenshots/test-results.jsonand analyze screenshots (SKILL.md). - Boundary markers: There are no delimiters or 'ignore' instructions provided to the agent for the content of the JSON file.
- Capability inventory: The skill has access to shell execution including
ls,grep,cat, and a custom script./qa-playwright-capture.sh(SKILL.md). - Sanitization: No sanitization or validation of the ingested JSON content is mentioned or implemented.
- [COMMAND_EXECUTION]: The skill executes shell commands and a local script to perform its auditing tasks.
- It uses
grep,ls, andcatto inspect the project structure and test results. - The execution of
./qa-playwright-capture.shwith a URL parameter (hardcoded ashttp://localhost:8000) presents a minor risk of command injection if the agent interpolates unvalidated user-supplied URLs into the command string.
Audit Metadata