MeMex Zero RAG

JPeetz4stars

The Karpathy LLM Wiki, production-ready. Zero-RAG personal knowledge base with MCP server, multi-agent support, knowledge graph visualization, and hallucination enforcement. No embeddings, no vector D

Knowledge Memory Productivity Autonomous Agents

Install Run on PlugMCP

GitHub Website

Run on PlugMCP

1# MeMex Zero RAG

README.mdMCP server

Transport: stdio
Status: Local

Power your agent

Add this server

Install from the publisher. Not a PlugMCP endpoint.

Run on PlugMCP

Host a copy with our keys and logs.

No packaged release in the registry yet. Follow the repository README.

RepositoryPublisher installStep-by-step: add this in Claude

Remote URL or a command on your machine? How the two installs differ

Run on PlugMCP

Host a copy with our keys and logs.

A new server you create. It is not this listing's official endpoint. Use the publisher install when you want their URL and their credentials.

Launch

Server facts

Local

Transport: stdio
Last checked: 2026-10-28

Stars: 4
Tools: —
Uptime: —

Overview

About MeMex Zero RAG

Tools

No tool list yet. Remote servers get one after the first health check.

FAQ

Questions

How do I install MeMex Zero RAG?

Copy the stdio config for Cursor, Claude Desktop, Claude Code, or VS Code on this page. It uses the publisher's install, not a PlugMCP proxy.

Does MeMex Zero RAG need an API key?

No required environment variable is listed on this record.

What transport does MeMex Zero RAG use?

MeMex Zero RAG runs locally over stdio.

When was MeMex Zero RAG last checked?

Last checked 2026-10-28. The Karpathy LLM Wiki, production-ready. Zero-RAG personal knowledge base with MCP server, multi-agent support, knowledge graph visualization, and hallucination enforcement. No embeddings, no vector DBs — just markdown + git.

From the publisher

Documentation

🧠 MeMex — Zero-RAG Personal Knowledge Base

The Karpathy LLM Wiki pattern, production-ready.

No embeddings. No vector databases. No infrastructure. Just markdown, git, and your LLM.

Quick Start • MCP Server • Prompts • Schema • Examples

"The human's job is to curate sources, direct the analysis, ask good questions, and think about what it all means. The LLM's job is everything else." — Andrej Karpathy

Based on Karpathy's LLM Wiki pattern, extended with:

🚫 Zero hallucination enforcement — every claim must cite a source
🤖 MCP server — expose your wiki to Claude Code, Codex, OpenClaw, Cursor
👥 Multi-agent support — multiple AI agents, one shared wiki, git handles conflicts
🔒 Human-in-the-loop conflict resolution — LLM flags contradictions, you decide truth
📊 Confidence tracking — per-claim certainty scores, quarantine mode for low-confidence pages
🌐 Knowledge graph — interactive force-directed visualization of all [[wikilinks]] (optional)

Why Not RAG?

Traditional RAG retrieves document chunks every query. Your LLM rediscovers knowledge from scratch. Every. Single. Time. Nothing compounds.

MeMex compiles knowledge once. Your LLM reads your sources and builds a structured, interlinked wiki. Every new source makes the whole wiki richer. Knowledge grows — it doesn't reset.

RAG:   Source → [retrieve] → LLM → Answer    (rediscovers every time)
MeMex: Source → [compile]  → Wiki → LLM       (compounds over time)

flowchart LR
    subgraph Input
        R["raw/ sources"]
    end
    
    subgraph Processing
        L["LLM reads & extracts"]
    end
    
    subgraph Knowledge
        W["wiki/ compiled knowledge"]
    end
    
    subgraph Output
        O["outputs/ reports"]
    end
    
    R --> L
    L --> W
    W --> O
    W -->|compounds over time| W

Key Principles

Principle	What it means
LLM-agnostic	Works with Claude, GPT, Gemini, Llama, Codex — any agent that reads/writes files
Zero infrastructure	No databases, no embeddings, no servers required. Markdown files + git
Git-native	Full version history, branching, rollback. Multi-agent via worktrees
Zero hallucination	Every claim cites a source. Unsourced = error. Lint enforces this on every PR
Human-in-the-loop	Conflicting sources get flagged and paused — you decide truth, not the LLM
Memory taxonomy	Optional `memory_type` labels (episodic/semantic/procedural) clarify how knowledge is used. Forward-only, never retro-downgrades. See SCHEMA.md

Quick Start

1. Clone

git clone https://github.com/JPeetz/MeMex-Zero-RAG.git my-wiki
cd my-wiki

2. Add a source

Drop any document into raw/:

cp ~/Downloads/interesting-article.md raw/

3. Tell your LLM to ingest it

Open Claude Code, Cursor, Codex, OpenClaw, or any AI coding agent and paste:

Read SCHEMA.md. Then ingest raw/interesting-article.md following the ingest workflow.

The LLM will:

Read the source and discuss key takeaways with you
Create a summary in wiki/sources/
Create or update entity and concept pages
Update wiki/index.md
Log the operation in wiki/log.md

4. Query your knowledge base

Read wiki/index.md. Based on the knowledge base, answer: [YOUR QUESTION].
Cite which wiki pages informed your answer.

5. Run the MCP server (optional)

Expose your wiki to Claude Code, OpenClaw, or any MCP-compatible agent:

pip install mcp
python mcp/server.py --wiki wiki/

# SSE for remote/multi-agent access
pip install uvicorn starlette sse-starlette
python mcp/server.py --transport sse --port 3001

That's it. No setup scripts. No API keys. No database migrations.

Directory Structure

your-wiki/
├── L1/                      # 🔒 Auto-loaded, git-ignored (private)
│   ├── identity.md          # Your preferences, context
│   ├── rules.md             # Hard constraints, gotchas
│   └── credentials.md       # API keys (NEVER committed)
│
├── raw/                     # 📥 Source documents (immutable)
│
├── wiki/                    # 📚 LLM-maintained knowledge base
│   ├── index.md             # Master catalog
│   ├── log.md               # Operation history
│   ├── contradictions.md    # Pending conflict resolutions
│   ├── sources/             # One summary per ingested source
│   ├── entities/            # People, orgs, tools, projects
│   ├── concepts/            # Ideas, frameworks, patterns
│   └── synthesis/           # Analyses, comparisons, insights
│
├── outputs/                 # 📤 Generated reports, slides
├── mcp/                     # 🔌 MCP server (stdio + SSE)
├── scripts/                 # CLI tools (memex ingest/search/lint)
├── SCHEMA.md                # 🧠 The brain — how the wiki works
├── PROMPTS.md               # 📋 Copy-paste prompts for all operations
└── README.md                # This file

The L1/L2 Architecture

Inspired by CPU cache hierarchy:

Layer	What	Loaded When	Contains
L1	`L1/` directory	Every session (auto-loaded by your agent)	Identity, rules, credentials
L2	`wiki/` directory	On-demand via queries	Deep knowledge, cross-references

L1 is git-ignored. It contains sensitive context that should never be committed.
L2 is the wiki. Versioned, shareable, grows over time.

Zero Hallucination Protocol

MeMex enforces citation discipline at every level:

Every claim must have a source: [Source: filename.md]
Lint catches violations: Unsourced claims are 🔴 ERROR, not warnings
Quarantine mode: Pages with >20% unsourced claims get status: quarantine
GitHub Actions: Broken links, missing citations, orphan pages checked on every PR

✅ Correct:  "Project X uses Redis [Source: architecture.md]"
❌ Rejected: "Project X uses Redis" (no source → lint error)

Conflict Resolution

When sources contradict each other:

LLM flags the conflict in wiki/contradictions.md
LLM stops and asks you which claim is authoritative
You decide — the LLM never auto-resolves truth
Decision is logged with rationale

MCP Tools

When running python mcp/server.py, your agent gets these tools:

Tool	Description
`wiki_search`	BM25 + semantic hybrid search across all pages
`wiki_read`	Read a specific page
`wiki_read_by_slug`	Read a page by bare slug (no type prefix); scans entities/concepts/sources/synthesis for a match
`wiki_list`	List pages by type (entities/concepts/sources/synthesis)
`wiki_query`	Natural language query with source attribution
`wiki_write`	Write a new page with git attribution (returns commit SHA); accepts optional lifecycle fields (`status`, `source_reliability_index`, `is_immutable`)
`wiki_flag`	Mark a page for revalidation — sets `status: quarantine` and records a reason/timestamp
`wiki_revalidate`	Close a revalidation cycle for a quarantined page: `confirm` (→ active), `update` (→ active with new content), or `retire` (→ stale)
`wiki_ingest`	Ingest a raw source into the wiki
`wiki_lint`	Run health checks on the wiki
`wiki_graph`	Generate wikilink relationship graph
`wiki_stats`	Summary statistics

Lifecycle management (`wiki_flag` / `wiki_revalidate` / `wiki_read_by_slug`)

MeMex pages carry a status field (draft | active | stale | quarantine) so agents and plugins (e.g. memex-hermes-plugin) can round-trip knowledge lifecycle state without a translation layer. See SCHEMA.md for the full field reference.

wiki_flag(page, reason) — quarantines a page (status: quarantine), stamping the reason and a UTC timestamp into frontmatter. Refuses to flag a page written with is_immutable: true.
wiki_revalidate(page, outcome, new_body?, new_sources?, rationale?) — resolves a quarantined page:
- outcome: confirm — clears quarantine, sets status: active, keeps existing content.
- outcome: update — replaces the body with new_body (required) and optionally appends new_sources citations, then sets status: active.
- outcome: retire — sets status: stale and clears flag metadata (reason/timestamp).
wiki_read_by_slug(slug) — looks up a page by bare slug across all four type directories (entities/, concepts/, sources/, synthesis/) so callers don't need to know a page's type prefix ahead of time. Returns {"found": false} when missing, or {"found": false, "ambiguous": true, "candidates": [...]} when the slug matches more than one type directory.

Claude Code config (`~/.claude/mcp.json`)

{
  "mcpServers": {
    "memex": {
      "command": "python",
      "args": ["mcp/server.py", "--wiki", "wiki/"]
    }
  }
}

OpenClaw config

{
  "plugins": {
    "mcp": {
      "servers": [{ "name": "memex", "transport": "stdio",
        "command": "python3 mcp/server.py --wiki wiki/" }]
    }
  }
}

Knowledge Graph Visualization

MeMex includes an optional knowledge graph that visualizes every [[wikilink]] in your wiki as an interactive, searchable force-directed graph.

# Enable when starting the MCP server
python3 mcp/server.py --enable-graph

The wiki_graph MCP tool builds graph/graph.html — open it in any browser. The graph auto-rebuilds on every wiki_write when enabled.

Feature	Detail
Nodes	Every wiki page
Edges	[[wikilinks]] between pages
Filters	By type (concept, entity, source, department…)
Search	Live search with highlight
Colors	Type-coded nodes with legend
Rendering	vis-network force-directed (dark theme)
Optional	Hidden unless `--enable-graph` is passed

Does not require --enable-graph to use the rest of MeMex — the graph is an entirely self-contained optional feature.

Multiple AI agents can write to the same wiki simultaneously using git worktrees:

# Each agent gets its own branch/worktree
git worktree add ../wiki-agent-1 feat/agent-1
git worktree add ../wiki-agent-2 feat/agent-2

# Agents commit with attribution
# wiki_write tool commits as: wiki(agent-name): add/update type/slug

Git handles authorship, history, and conflict detection natively. See GUIDE.md for the full multi-agent setup.

CLI Tooling

# Add to PATH
export PATH="$PATH:/path/to/memex/scripts"

memex ingest paper.pdf           # PDF with academic metadata
memex ingest https://example.com # Web page
memex search "machine learning"  # BM25 + semantic search
memex lint --fix                 # Check wiki health
memex graph                      # Generate wikilink graph
memex serve                      # Start MCP server

Requirements

Any AI coding agent: Claude Code, Cursor, Codex, OpenClaw, Gemini CLI, etc.
Git for version control
Python 3.9+ for MCP server (optional)
That's it — no databases, no APIs, no infrastructure

Optional Enhancements

Tool	What it does	When to add
Obsidian	Graph view, backlinks, visual navigation	From day one
qmd	Local hybrid search for large wikis	100+ pages
Marp	Generate slide decks from wiki content	When presenting
Batch API	50% cost reduction for bulk ingests	Large ingest sessions

Credits

Vannevar Bush — Original Memex concept (1945)
Andrej Karpathy — LLM Wiki pattern
MehmetGoekce/llm-wiki — L1/L2 cache architecture inspiration
rohitg00's LLM Wiki v2 — Lifecycle and consolidation patterns

Author

Joerg Peetz — with contributions from Molty, Coconut, Marvin (AI collaborators on the MeMex team)

License

MIT — Use it, fork it, adapt it, share it.

Start small. Ingest one source. Ask one question. Watch the wiki grow.

Part of the AgentForge Ecosystem

MeMex Zero RAG

JPeetz4stars

Knowledge Memory Productivity Autonomous Agents

Install Run on PlugMCP

GitHub Website

Run on PlugMCP

1# MeMex Zero RAG

README.mdMCP server

Transport: stdio
Status: Local

Power your agent

Add this server

Install from the publisher. Not a PlugMCP endpoint.

Run on PlugMCP

Host a copy with our keys and logs.

No packaged release in the registry yet. Follow the repository README.

RepositoryPublisher installStep-by-step: add this in Claude

Remote URL or a command on your machine? How the two installs differ

Run on PlugMCP

Host a copy with our keys and logs.

A new server you create. It is not this listing's official endpoint. Use the publisher install when you want their URL and their credentials.

Launch

Server facts

Local

Transport: stdio
Last checked: 2026-10-28

Stars: 4
Tools: —
Uptime: —

Overview

About MeMex Zero RAG

Tools

No tool list yet. Remote servers get one after the first health check.

FAQ

Questions

How do I install MeMex Zero RAG?

Copy the stdio config for Cursor, Claude Desktop, Claude Code, or VS Code on this page. It uses the publisher's install, not a PlugMCP proxy.

Does MeMex Zero RAG need an API key?

No required environment variable is listed on this record.

What transport does MeMex Zero RAG use?

MeMex Zero RAG runs locally over stdio.

When was MeMex Zero RAG last checked?

From the publisher

Documentation

🧠 MeMex — Zero-RAG Personal Knowledge Base

The Karpathy LLM Wiki pattern, production-ready.

No embeddings. No vector databases. No infrastructure. Just markdown, git, and your LLM.

Quick Start • MCP Server • Prompts • Schema • Examples

"The human's job is to curate sources, direct the analysis, ask good questions, and think about what it all means. The LLM's job is everything else." — Andrej Karpathy

Based on Karpathy's LLM Wiki pattern, extended with:

🚫 Zero hallucination enforcement — every claim must cite a source
🤖 MCP server — expose your wiki to Claude Code, Codex, OpenClaw, Cursor
👥 Multi-agent support — multiple AI agents, one shared wiki, git handles conflicts
🔒 Human-in-the-loop conflict resolution — LLM flags contradictions, you decide truth
📊 Confidence tracking — per-claim certainty scores, quarantine mode for low-confidence pages
🌐 Knowledge graph — interactive force-directed visualization of all [[wikilinks]] (optional)

Why Not RAG?

Traditional RAG retrieves document chunks every query. Your LLM rediscovers knowledge from scratch. Every. Single. Time. Nothing compounds.

MeMex compiles knowledge once. Your LLM reads your sources and builds a structured, interlinked wiki. Every new source makes the whole wiki richer. Knowledge grows — it doesn't reset.

RAG:   Source → [retrieve] → LLM → Answer    (rediscovers every time)
MeMex: Source → [compile]  → Wiki → LLM       (compounds over time)

flowchart LR
    subgraph Input
        R["raw/ sources"]
    end
    
    subgraph Processing
        L["LLM reads & extracts"]
    end
    
    subgraph Knowledge
        W["wiki/ compiled knowledge"]
    end
    
    subgraph Output
        O["outputs/ reports"]
    end
    
    R --> L
    L --> W
    W --> O
    W -->|compounds over time| W

Key Principles

Principle	What it means
LLM-agnostic	Works with Claude, GPT, Gemini, Llama, Codex — any agent that reads/writes files
Zero infrastructure	No databases, no embeddings, no servers required. Markdown files + git
Git-native	Full version history, branching, rollback. Multi-agent via worktrees
Zero hallucination	Every claim cites a source. Unsourced = error. Lint enforces this on every PR
Human-in-the-loop	Conflicting sources get flagged and paused — you decide truth, not the LLM
Memory taxonomy	Optional `memory_type` labels (episodic/semantic/procedural) clarify how knowledge is used. Forward-only, never retro-downgrades. See SCHEMA.md

Quick Start

1. Clone

git clone https://github.com/JPeetz/MeMex-Zero-RAG.git my-wiki
cd my-wiki

2. Add a source

Drop any document into raw/:

cp ~/Downloads/interesting-article.md raw/

3. Tell your LLM to ingest it

Open Claude Code, Cursor, Codex, OpenClaw, or any AI coding agent and paste:

Read SCHEMA.md. Then ingest raw/interesting-article.md following the ingest workflow.

The LLM will:

Read the source and discuss key takeaways with you
Create a summary in wiki/sources/
Create or update entity and concept pages
Update wiki/index.md
Log the operation in wiki/log.md

4. Query your knowledge base

Read wiki/index.md. Based on the knowledge base, answer: [YOUR QUESTION].
Cite which wiki pages informed your answer.

5. Run the MCP server (optional)

Expose your wiki to Claude Code, OpenClaw, or any MCP-compatible agent:

pip install mcp
python mcp/server.py --wiki wiki/

# SSE for remote/multi-agent access
pip install uvicorn starlette sse-starlette
python mcp/server.py --transport sse --port 3001

That's it. No setup scripts. No API keys. No database migrations.

Directory Structure

your-wiki/
├── L1/                      # 🔒 Auto-loaded, git-ignored (private)
│   ├── identity.md          # Your preferences, context
│   ├── rules.md             # Hard constraints, gotchas
│   └── credentials.md       # API keys (NEVER committed)
│
├── raw/                     # 📥 Source documents (immutable)
│
├── wiki/                    # 📚 LLM-maintained knowledge base
│   ├── index.md             # Master catalog
│   ├── log.md               # Operation history
│   ├── contradictions.md    # Pending conflict resolutions
│   ├── sources/             # One summary per ingested source
│   ├── entities/            # People, orgs, tools, projects
│   ├── concepts/            # Ideas, frameworks, patterns
│   └── synthesis/           # Analyses, comparisons, insights
│
├── outputs/                 # 📤 Generated reports, slides
├── mcp/                     # 🔌 MCP server (stdio + SSE)
├── scripts/                 # CLI tools (memex ingest/search/lint)
├── SCHEMA.md                # 🧠 The brain — how the wiki works
├── PROMPTS.md               # 📋 Copy-paste prompts for all operations
└── README.md                # This file

The L1/L2 Architecture

Inspired by CPU cache hierarchy:

Layer	What	Loaded When	Contains
L1	`L1/` directory	Every session (auto-loaded by your agent)	Identity, rules, credentials
L2	`wiki/` directory	On-demand via queries	Deep knowledge, cross-references

L1 is git-ignored. It contains sensitive context that should never be committed.
L2 is the wiki. Versioned, shareable, grows over time.

Zero Hallucination Protocol

MeMex enforces citation discipline at every level:

Every claim must have a source: [Source: filename.md]
Lint catches violations: Unsourced claims are 🔴 ERROR, not warnings
Quarantine mode: Pages with >20% unsourced claims get status: quarantine
GitHub Actions: Broken links, missing citations, orphan pages checked on every PR

✅ Correct:  "Project X uses Redis [Source: architecture.md]"
❌ Rejected: "Project X uses Redis" (no source → lint error)

Conflict Resolution

When sources contradict each other:

LLM flags the conflict in wiki/contradictions.md
LLM stops and asks you which claim is authoritative
You decide — the LLM never auto-resolves truth
Decision is logged with rationale

MCP Tools

When running python mcp/server.py, your agent gets these tools:

Tool	Description
`wiki_search`	BM25 + semantic hybrid search across all pages
`wiki_read`	Read a specific page
`wiki_read_by_slug`	Read a page by bare slug (no type prefix); scans entities/concepts/sources/synthesis for a match
`wiki_list`	List pages by type (entities/concepts/sources/synthesis)
`wiki_query`	Natural language query with source attribution
`wiki_write`	Write a new page with git attribution (returns commit SHA); accepts optional lifecycle fields (`status`, `source_reliability_index`, `is_immutable`)
`wiki_flag`	Mark a page for revalidation — sets `status: quarantine` and records a reason/timestamp
`wiki_revalidate`	Close a revalidation cycle for a quarantined page: `confirm` (→ active), `update` (→ active with new content), or `retire` (→ stale)
`wiki_ingest`	Ingest a raw source into the wiki
`wiki_lint`	Run health checks on the wiki
`wiki_graph`	Generate wikilink relationship graph
`wiki_stats`	Summary statistics

Lifecycle management (`wiki_flag` / `wiki_revalidate` / `wiki_read_by_slug`)

wiki_flag(page, reason) — quarantines a page (status: quarantine), stamping the reason and a UTC timestamp into frontmatter. Refuses to flag a page written with is_immutable: true.
wiki_revalidate(page, outcome, new_body?, new_sources?, rationale?) — resolves a quarantined page:
- outcome: confirm — clears quarantine, sets status: active, keeps existing content.
- outcome: update — replaces the body with new_body (required) and optionally appends new_sources citations, then sets status: active.
- outcome: retire — sets status: stale and clears flag metadata (reason/timestamp).
wiki_read_by_slug(slug) — looks up a page by bare slug across all four type directories (entities/, concepts/, sources/, synthesis/) so callers don't need to know a page's type prefix ahead of time. Returns {"found": false} when missing, or {"found": false, "ambiguous": true, "candidates": [...]} when the slug matches more than one type directory.

Claude Code config (`~/.claude/mcp.json`)

{
  "mcpServers": {
    "memex": {
      "command": "python",
      "args": ["mcp/server.py", "--wiki", "wiki/"]
    }
  }
}

OpenClaw config

{
  "plugins": {
    "mcp": {
      "servers": [{ "name": "memex", "transport": "stdio",
        "command": "python3 mcp/server.py --wiki wiki/" }]
    }
  }
}

Knowledge Graph Visualization

MeMex includes an optional knowledge graph that visualizes every [[wikilink]] in your wiki as an interactive, searchable force-directed graph.

# Enable when starting the MCP server
python3 mcp/server.py --enable-graph

The wiki_graph MCP tool builds graph/graph.html — open it in any browser. The graph auto-rebuilds on every wiki_write when enabled.

Feature	Detail
Nodes	Every wiki page
Edges	[[wikilinks]] between pages
Filters	By type (concept, entity, source, department…)
Search	Live search with highlight
Colors	Type-coded nodes with legend
Rendering	vis-network force-directed (dark theme)
Optional	Hidden unless `--enable-graph` is passed

Does not require --enable-graph to use the rest of MeMex — the graph is an entirely self-contained optional feature.

Multiple AI agents can write to the same wiki simultaneously using git worktrees:

# Each agent gets its own branch/worktree
git worktree add ../wiki-agent-1 feat/agent-1
git worktree add ../wiki-agent-2 feat/agent-2

# Agents commit with attribution
# wiki_write tool commits as: wiki(agent-name): add/update type/slug

Git handles authorship, history, and conflict detection natively. See GUIDE.md for the full multi-agent setup.

CLI Tooling

# Add to PATH
export PATH="$PATH:/path/to/memex/scripts"

memex ingest paper.pdf           # PDF with academic metadata
memex ingest https://example.com # Web page
memex search "machine learning"  # BM25 + semantic search
memex lint --fix                 # Check wiki health
memex graph                      # Generate wikilink graph
memex serve                      # Start MCP server

Requirements

Any AI coding agent: Claude Code, Cursor, Codex, OpenClaw, Gemini CLI, etc.
Git for version control
Python 3.9+ for MCP server (optional)
That's it — no databases, no APIs, no infrastructure

Optional Enhancements

Tool	What it does	When to add
Obsidian	Graph view, backlinks, visual navigation	From day one
qmd	Local hybrid search for large wikis	100+ pages
Marp	Generate slide decks from wiki content	When presenting
Batch API	50% cost reduction for bulk ingests	Large ingest sessions

Credits

Vannevar Bush — Original Memex concept (1945)
Andrej Karpathy — LLM Wiki pattern
MehmetGoekce/llm-wiki — L1/L2 cache architecture inspiration
rohitg00's LLM Wiki v2 — Lifecycle and consolidation patterns

Author

Joerg Peetz — with contributions from Molty, Coconut, Marvin (AI collaborators on the MeMex team)

License

MIT — Use it, fork it, adapt it, share it.

Start small. Ingest one source. Ask one question. Watch the wiki grow.

Part of the AgentForge Ecosystem

Add this server

Host a copy with our keys and logs.

Server facts

About MeMex Zero RAG

Tools

Questions

How do I install MeMex Zero RAG?

Does MeMex Zero RAG need an API key?

What transport does MeMex Zero RAG use?

When was MeMex Zero RAG last checked?

Documentation

🧠 MeMex — Zero-RAG Personal Knowledge Base

Why Not RAG?

Key Principles

Quick Start

1. Clone

2. Add a source

3. Tell your LLM to ingest it

4. Query your knowledge base

5. Run the MCP server (optional)

Directory Structure

The L1/L2 Architecture

Zero Hallucination Protocol

Conflict Resolution

MCP Tools

Lifecycle management (wiki_flag / wiki_revalidate / wiki_read_by_slug)

Claude Code config (~/.claude/mcp.json)

OpenClaw config

Knowledge Graph Visualization

CLI Tooling

Requirements

Optional Enhancements

Credits

Author

License

Add this server

Host a copy with our keys and logs.

Server facts

About MeMex Zero RAG

Tools

Questions

How do I install MeMex Zero RAG?

Does MeMex Zero RAG need an API key?

What transport does MeMex Zero RAG use?

When was MeMex Zero RAG last checked?

Documentation

🧠 MeMex — Zero-RAG Personal Knowledge Base

Why Not RAG?

Key Principles

Quick Start

1. Clone

2. Add a source

3. Tell your LLM to ingest it

4. Query your knowledge base

5. Run the MCP server (optional)

Directory Structure

The L1/L2 Architecture

Zero Hallucination Protocol

Conflict Resolution

MCP Tools

Lifecycle management (wiki_flag / wiki_revalidate / wiki_read_by_slug)

Claude Code config (~/.claude/mcp.json)

OpenClaw config

Knowledge Graph Visualization

CLI Tooling

Requirements

Optional Enhancements

Credits

Author

License

Lifecycle management (`wiki_flag` / `wiki_revalidate` / `wiki_read_by_slug`)

Claude Code config (`~/.claude/mcp.json`)

Lifecycle management (`wiki_flag` / `wiki_revalidate` / `wiki_read_by_slug`)

Claude Code config (`~/.claude/mcp.json`)