Kogcat Mcp
Runs a local calibration engine that surfaces counterexamples and blind spots from your own knowledge base before you act on an AI answer. Exposes MCP tools for search, node retrieval, edge traversal, and calibration review. The engine downloads once (around 130 MB total) and runs as a background sidecar shared across any MCP client on your machine. Fires automatically on judgment questions when used as a Claude Code or Codex plugin, otherwise your model calls the tools directly. Requires uv, supports macOS Apple Silicon and Windows x86_64. Your knowledge base and calibration stay local. Reach for it when the cost of a wrong call on strategy, tradeoffs, or decisions outweighs the friction of a second opinion.
- tools
- 5
- transport
- STDIO
- last updated
- Jun 13, 2026
Tools
5 toolsPublic tool metadata for what this MCP can expose to an agent.
searchSearches the knowledge base.
Searches the knowledge base.
No parameter schema in public metadata yet.
node
No parameter schema in public metadata yet.
edges
No parameter schema in public metadata yet.
calibrate
No parameter schema in public metadata yet.
calibrate_review
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageReadme
KogCat
AI makes the answer smoother. KogCat makes the judgment sound.
English | 中文 · Website: https://www.kogcat.com
A local-first judgment calibration layer for Claude Code and Codex. Before you act on an AI answer, it surfaces the counterexamples, the boundaries, the blind spots — drawn from a knowledge base that lives on your machine. It won't replace your model. It won't slow you down. The call stays yours.
See the difference
"I read 30 minutes a day but nothing sticks. Should I take more detailed notes?"
Plain AI — Try the Cornell method, highlight key passages, add Anki for spaced repetition.
KogCat — More notes will likely make it worse. The bottleneck isn't capture. It's retrieval. Your knowledge base holds a claim you marked high-confidence: the "I get it" feeling while re-reading is the least reliable signal of real recall. So try this once — finish a section, close the book, write what you remember. Then compare it to what you thought you had.
When it speaks up
Details
- Repository
- github.com/kogcat/cc-kogcat
- Category
- Documents & Knowledge
- Package
kogcat-mcp- Transport
- STDIO
- Registry
- MCP registry, active
Related Documents & Knowledge MCP servers
View all
Laddro Careerladdro-app/laddro-career-mcpTailor resumes, generate cover letters, render CVs as PDF, and browse 22+ templates.0stars
md2signsteamdiesel/md2sign-claude-pluginSend markdown contracts for legally binding e-signatures. Draft, preview, send and track signing.0stars
Deweymeetdewey/mcpAgentic search over your Dewey document collections from any MCP-compatible client.0stars
pdfverapdfvera/pdfvera-mcpPDF invoices to XRechnung/ZUGFeRD; Brazilian bank statement PDFs to OFX, CSV and Excel.0stars
Hermes — Air Quality Intelligencecom.southlondonscientific/hermesUK air quality MCP: live readings, historical trends, LAQM stats, WHO checks, knowledge base.0stars
Taxact Mcpcom.taxact/taxact-mcpTax filing guidance, cost comparison, document checklists, and refund estimates by TaxAct.0stars