secondmem Agent Constitution

Ingest, organize, and rebalance local Markdown knowledge bases for agent consumption.

9|Updated Apr 4, 2026
One-click install
npx skills add https://github.com/om-ashish-soni/2nd-mem --skill secondmem-agent-constitution
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: secondmem Agent Constitution
Source: https://github.com/om-ashish-soni/2nd-mem/tree/main/.secondmem
Command: npx skills add https://github.com/om-ashish-soni/2nd-mem --skill secondmem-agent-constitution

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The secondmem agent provides a deterministic framework for ingesting, structuring, and maintaining a local Markdown-based knowledge base so that agents can navigate and reason without external databases.

Core Features & Use Cases

  • Ingest: parse input content, classify, store, and rebalance to preserve hierarchy consistency.
  • Organize: enforce a kebab-case naming convention, hierarchical 'hierarchy.md' contracts, and cross-reference links.
  • Query: support agent-oriented retrieval and synthesis over a navigable local wiki.
  • Use Case: maintain a living local memory for a single-user AI assistant that can answer questions and update knowledge structures offline.

Quick Start

Ingest a new note or document and trigger a rebalance to refresh the knowledge graph.

Frequently Asked Questions about secondmem Agent Constitution

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

FAQPage Schema
How do I organize a local Markdown knowledge base for an AI assistant?

You can organize a local Markdown knowledge base by ingesting content and triggering a rebalance to enforce a kebab-case naming convention and maintain hierarchical integrity. This creates a navigable local wiki for agent consumption.

How does cross-reference maintenance work when ingesting Markdown files?

Cross-reference maintenance works by parsing input content, classifying it, and storing it while automatically rebalancing the knowledge graph. This ensures hierarchical consistency and valid links across the local Markdown wiki structure.

Can I use a Markdown knowledge base offline with a single-user AI assistant?

Yes, you can maintain a living local memory for a single-user AI assistant that operates entirely offline. The system supports agent-oriented retrieval and synthesis over a structured hierarchy of Markdown files without external databases.

What is the best way to structure private notes for agent consumption?

The best way to structure private notes is applying deterministic ingestion rules that enforce hierarchical contracts and cross-reference links. This framework allows agents to navigate and reason over a local Markdown wiki without external dependencies.

Do I need an external database to query a local Markdown wiki?

No, you do not need an external database to query a local Markdown wiki. The framework provides deterministic agent-oriented retrieval and synthesis directly over structured Markdown files using local hierarchy contracts.