manual-testing

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill directs the agent to execute various shell commands for building, testing, and verifying the environment (e.g., npm run build, cargo test, curl, lsof). These operations are standard for a development and QA assistant.
  • [INDIRECT_PROMPT_INJECTION]: The skill identifies testing requirements by analyzing recent file changes and conversation history. This creates a surface where malicious instructions embedded in project files could potentially influence the agent's behavior during the testing process.
  • Ingestion points: Workflow Step 1 in SKILL.md ("Review recent file changes and conversation history").
  • Boundary markers: None identified in the instructions.
  • Capability inventory: Execution of build tools, test runners, and network utilities (npm, cargo, curl, etc.).
  • Sanitization: No explicit sanitization or validation of the content found in project files is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 01:25 PM
Security Audit — agent-trust-hub — manual-testing