workflow-spec-sync

Pass

Audited by Gen Agent Trust Hub on Nov 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to run a local Node.js script (node .claude/scripts/lib/workflow-skill-contract.cjs --write) to maintain the integrity of its own documentation block.
  • [DYNAMIC_EXECUTION]: The skill manages complex workflows by dynamically loading instructions from external SKILL.md files and utilizing spawn_agent to delegate work to sub-agents.
  • [INDIRECT_PROMPT_INJECTION]: The skill reads project-level configuration files (docs/project-config.json, .claude/workflows.json) to determine its logic, which represents an attack surface for indirect prompt injection.
  • Ingestion points: docs/project-config.json, .claude/workflows.json, docs/project-reference/spec-principles.md, and user prompts.
  • Boundary markers: The skill provides no specific delimiters or instructions to ignore potential commands within these files.
  • Capability inventory: The skill can execute local Node.js scripts, create sub-agents, and perform filesystem operations.
  • Sanitization: There are no requirements for the agent to sanitize or validate the content of ingested configuration files.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 5, 2026, 06:42 PM