mantis-architecture

Synthesize codebase learnings into an interlinked Markdown Knowledge Base.

731|84|Updated Jun 15, 2026
One-click install
npx skills add https://github.com/google/mantis --skill mantis-architecture
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mantis-architecture
Source: https://github.com/google/mantis/tree/main/mantis-architecture
Command: npx skills add https://github.com/google/mantis --skill mantis-architecture

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the problem of fragmented knowledge by synthesizing raw, ephemeral findings into a structured, interlinked Markdown Knowledge Base that serves as a persistent memory for security agents.

Core Features & Use Cases

  • Knowledge Synthesis: Automatically translates raw learnings and structural analysis into a canonical architecture, entity, and vulnerability map.
  • Drift Prevention: Uses snapshot-based freshness gates to ensure the Knowledge Base remains synchronized with the codebase, preventing reliance on stale security assertions.
  • Use Case: Use this at the start of a security review loop to build a comprehensive architectural map of a new codebase, ensuring all subsequent vulnerability discovery is grounded in an accurate, up-to-date system model.

Quick Start

Invoke the mantis-architecture skill to build or update the project knowledge base using the current workspace state.

Frequently Asked Questions about mantis-architecture

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I synthesize codebase analysis into a knowledge base for security reviews?

To synthesize codebase analysis into a knowledge base, this skill translates raw structural learnings and vulnerability patterns into an interlinked Markdown map. It builds a persistent architectural memory to ground security agents during review loops.

What is the best way to map system architecture and components for vulnerability discovery?

Mapping system architecture for vulnerability discovery is best handled by synthesizing raw codebase structural analysis into a canonical, interlinked Markdown knowledge base. This provides an accurate, persistent system model for security agents.

How do I prevent stale security assertions when analyzing codebase architecture?

Preventing stale security assertions requires state-managed snapshot provenance to gate knowledge base freshness. This ensures the architectural map remains synchronized with the current codebase state during security reviews.

Can I use this architecture synthesis skill at the start of a security review loop?

Yes, you can use this architecture synthesis skill at the start of a security review loop. It builds or updates the project knowledge base using the current workspace state to ensure subsequent vulnerability discovery is accurately grounded.

Why does my security agent rely on fragmented knowledge instead of a unified architectural model?

Security agents rely on fragmented knowledge when lacking a persistent memory layer. This skill solves that by synthesizing ephemeral findings into a structured, interlinked Markdown knowledge base for canonical architecture and vulnerability mapping.