antigravity-workflows

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill loads orchestration logic from docs/WORKFLOWS.md and data/workflows.json (referenced in SKILL.md), creating a surface for embedded instructions to influence the agent.
  • Ingestion points: Documented in SKILL.md, the agent reads local documentation and data files to define its task list.
  • Boundary markers: The skill does not provide delimiters to separate workflow data from system instructions.
  • Capability inventory: The skill can invoke sub-skills for file writes and network operations as detailed in SKILL.md and resources/implementation-playbook.md.
  • Sanitization: No validation of workflow files is specified in the implementation playbook.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 10:50 AM
Security Audit — agent-trust-hub — antigravity-workflows