workflow-integration-test

Pass

Audited by Gen Agent Trust Hub on Nov 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill operates as a high-level orchestrator that processes external data from user prompts, workflow registries, and project documentation while possessing capabilities to execute commands and spawn sub-agents.
  • Ingestion points: User-provided prompts, the workflow registry (.claude/workflows.json), project specifications, and test source code.
  • Boundary markers: Workflows are managed through the $start-workflow protocol and a structured task tracking system to separate execution phases.
  • Capability inventory: Spawns sub-agents (spawn_agent), executes shell commands via referenced skills (e.g., $integration-test), and modifies project files.
  • Sanitization: No explicit sanitization or escaping mechanisms for the ingested data are documented within the skill files.
  • [COMMAND_EXECUTION]: The skill directs the agent to run a local Node.js script (.claude/scripts/lib/workflow-skill-contract.cjs) to regenerate internal contract blocks within the documentation.
  • [EXTERNAL_DOWNLOADS]: The skill references a 'Python readiness' policy intended to repair or provision missing Python runtimes on Windows, macOS, and Linux to ensure the environment is capable of running integration tests.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 9, 2026, 01:34 AM