specforge-verify

Pass

Audited by Gen Agent Trust Hub on May 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local Node.js scripts (.specforge/tools/create-artifact.mjs and .specforge/tools/gate.mjs) and performs arbitrary shell commands for testing, building, and linting as part of its core verification logic.
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface:
  • Ingestion points: Reads and extracts verification strategies from external design and tasks files.
  • Boundary markers: None; the instructions do not specify delimiters or warnings to ignore embedded instructions within the processed data.
  • Capability inventory: Significant; can execute shell commands (npm test, make, etc.) and Node.js scripts with local file system access.
  • Sanitization: None; the skill instructs the agent to run commands directly based on the extracted strategy.
Audit Metadata
Risk Level
SAFE
Analyzed
May 12, 2026, 10:43 AM
Security Audit — agent-trust-hub — specforge-verify