skills/howells/arc/testing/Gen Agent Trust Hub

testing

Pass

Audited by Gen Agent Trust Hub on May 10, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface (Category 8) due to how it handles project-specific data and configuration files.\n
  • Ingestion points: The skill reads external data from the user's repository via .ruler/testing.md and detects/reads testing framework configurations (e.g., vitest.config.*, playwright.config.*).\n
  • Boundary markers: The prompt templates used to dispatch tasks to sub-agents (e.g., unit-test-writer, e2e-test-writer) use simple placeholders like [paste unit test cases] or [endpoints to mock] without employing XML tags or clear delimiters to isolate untrusted content from the system instructions.\n
  • Capability inventory: The agents have the capability to write files to the repository and execute shell commands (e.g., pnpm vitest run).\n
  • Sanitization: No sanitization, validation, or escaping of the ingested configuration or code is performed before it is interpolated into the agent tasks.\n- [COMMAND_EXECUTION]: The skill facilitates the execution of pnpm vitest run to verify test suites. This behavior is documented and aligns with the skill's primary purpose as a testing tool.
Audit Metadata
Risk Level
SAFE
Analyzed
May 10, 2026, 04:39 PM
Security Audit — agent-trust-hub — testing