mantis-threat-model

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFE
Full Analysis
  • File System Operations: The skill manages a structured Knowledge Base (KB) within a workspace/ directory. It reads architectural definitions and entity files to generate a THREAT_MODEL.md file. These operations are restricted to the workspace and archive directories, following a 'findings-only' approach that avoids direct interaction with source code.
  • Provenance and State Management: It utilizes a .mantis_state.json file to track pass numbers and snapshot IDs. This ensures that the generated threat model is correctly associated with a specific state of the system architecture.
  • Deployment Intent Guardrails: The instructions include a specific 'PRODUCTION-SIGNAL' checklist. This is a policy-driven mechanism used to determine the criticality of the findings based on whether the analyzed system is production-ready or a sample/test project.
  • Data Ingestion Considerations: The skill processes files generated in previous stages of a pipeline. While it ingests external data (architectural descriptions), its functionality is limited to synthesizing structured Markdown reports. It does not execute the content of the files it reads.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 05:43 PM
Security Audit — agent-trust-hub — mantis-threat-model