workflow-seed-test-data

Pass

Audited by Gen Agent Trust Hub on Nov 10, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes project-specific configuration and documentation files to guide the generation of test seeders, which creates a vulnerability surface for instructions embedded in those files.
  • Ingestion points: Reads data from docs/project-config.json, seed-test-data-reference.md, and .claude/workflows.json to extract project context and workflow definitions.
  • Boundary markers: No explicit boundary markers or instructions to ignore embedded prompts are specified when the agent reads these project files.
  • Capability inventory: The skill can execute local scripts (Node.js), write files to the tmp/reports/ directory, call multiple other agent skills, and spawn sub-agents.
  • Sanitization: There is no evidence of validation or sanitization of the ingested project content before it is used to influence agent behavior.
  • [COMMAND_EXECUTION]: The skill directs the execution of a local Node.js script for maintenance purposes.
  • Evidence: Instructs the agent or user to run node .claude/scripts/lib/workflow-skill-contract.cjs --write to regenerate the workflow calls block in the documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 10, 2026, 08:16 PM