knowledge-ops

What problem does it solve?

Manage fragmented knowledge scattered across local files, conversational memory, vector stores, and code repositories by providing a single operational workflow to ingest, deduplicate, organize, and retrieve durable context across systems.

Core Features & Use Cases

• Multi-layer ingestion: Classify and route content into the appropriate layer (quick memory files, MCP graph, KB repo, external DB) based on type and durability needs.
• Deduplication and indexing: Search existing memory and MCP nodes before creating new entries, update existing records, and refresh indexes and summaries.
• Sync and commit workflows: Mirror active execution truth to GitHub/Linear when appropriate, redact sensitive data, and commit synthesized session exports or long-form notes to a knowledge base repo.
• Use Case: After a technical design review, extract the conversation, deduplicate against project memory, update MCP relations, write a short summary in the project memory file, and commit the synthesized notes to the KB repo.

Quick Start

Ingest and sync selected documents and session notes into the knowledge base while updating Claude memory and the MCP memory graph and committing summarized artifacts to the KB repository.