Skip to main content
Glama
Project Synapse
by angrysky56

🧠 Project Synapse MCP Server

Autonomous Knowledge Synthesis Engine with LLM-WIKI Integration

Documentation

Project Synapse is an MCP (Model Context Protocol) server that combines a Neo4j 2026.x graph database with an Obsidian Markdown wiki to create a persistent, compounding knowledge base. Raw text is processed through a semantic pipeline into interconnected graph nodes with vector embeddings, while a human-readable wiki layer provides browsable, interlinked Markdown pages.

πŸ“š Documentation

For detailed information on setting up and using Project Synapse, please refer to the following guides:

Related MCP server: KG Memory

What This Is (and Isn't)

This is a knowledge system, not a code editor. It's for the thinking, research, and writing that surrounds projects β€” architecture decisions, domain research, design rationale, reference material, meeting notes.

Code lives in its repo. Knowledge about the code lives here.

Use cases:

  • Research deep-dives that accumulate over weeks/months

  • Project knowledge bases (why decisions were made, not just what)

  • Personal knowledge management (articles, books, podcast notes)

  • Collaborative brainstorming with AI as the wiki maintainer

Per-project setup: Create a separate Obsidian vault + GitHub repo for each project. Point the WIKI_VAULT_PATH env var at it. One Neo4j instance can serve multiple projects (graphs coexist).

Architecture

Web / Raw Sources
         β”‚
    [defuddle]          ← cleans web content before ingestion
         β”‚
         β–Ό
β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”     β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚  Semantic Pipeline   │────▢│  Neo4j Knowledge    β”‚
β”‚  (Montague Grammar,  β”‚     β”‚  Graph (entities,   β”‚
β”‚   NLP, embeddings)   β”‚     β”‚  facts, vectors)    β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜     β””β”€β”€β”€β”€β”€β”€β”€β”€β”¬β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜
                                      β”‚
                              β”Œβ”€β”€β”€β”€β”€β”€β”€β”΄β”€β”€β”€β”€β”€β”€β”€β”
                              β”‚ Wiki Adapter  β”‚
                              β””β”€β”€β”€β”€β”€β”€β”€β”¬β”€β”€β”€β”€β”€β”€β”€β”˜
                                      β”‚
                              β”Œβ”€β”€β”€β”€β”€β”€β”€β–Όβ”€β”€β”€β”€β”€β”€β”€β”
                              β”‚ Obsidian Vaultβ”‚
                              β”‚ (Markdown,    β”‚
                              β”‚  Git-synced)  β”‚
                              β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜

Key Features

Knowledge Graph (Neo4j 2026.x)

  • Native VECTOR type with ANN semantic search

  • Fulltext BM25 indexes for keyword search

  • Hybrid search (vector + BM25 score fusion)

  • Graph traversal for discovering hidden relationships

  • Montague Grammar parser for formal semantic analysis

  • Hybrid Extraction Pipeline: Merges LLM-based extraction (Gemma 2 9b via Ollama) with spaCy NER for high-precision entity and relationship discovery

  • Zettelkasten engine for autonomous insight generation

LLM-WIKI Integration

  • Bridges Obsidian Markdown vault with the Neo4j graph

  • Full page CRUD with YAML frontmatter

  • Automatic index generation and append-only log

  • Health checks: orphan detection, broken wikilinks, missing frontmatter

  • Delta-sync manifest (content hashing) for efficient graph sync

  • Based on Andrej Karpathy's LLM Wiki pattern

Web Content Ingestion (defuddle)

  • wiki_fetch_url fetches any URL, strips navigation/ads/clutter via defuddle, ingests into Neo4j, and archives to Clippings/ β€” one call, fully automated

  • wiki_ingest_raw auto-moves processed files from raw/ to Clippings/ β€” inbox stays clean

  • raw/ is a true inbox: empty after every session

Local-Only Embeddings (No Paid APIs)

  • sentence-transformers (default) β€” runs on GPU

  • Ollama (optional) β€” any local embedding model

  • All vectors stored natively in Neo4j via db.create.setNodeVectorProperty()

Quick Start

Prerequisites

  • Python 3.12+

  • Neo4j 2026.x (Community or Enterprise)

  • uv package manager (pip install uv)

  • Obsidian with the Git community plugin

  • A GitHub repo for the wiki vault (can be private)

  • Node.js + defuddle (for web content fetching β€” see below)

Neo4j Setup

# Ubuntu/Debian β€” see neo4j.com for other platforms
sudo apt install neo4j
sudo systemctl start neo4j
sudo systemctl enable neo4j
# Set password (default user: neo4j)
sudo neo4j-admin set-initial-password your_password

defuddle Setup

defuddle extracts clean markdown from web pages, stripping navigation, ads, and boilerplate before ingestion. Required for wiki_fetch_url.

# Install Node.js if not present (via nvm recommended)
curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.39.0/install.sh | bash
nvm install --lts
nvm use --lts

# Install defuddle globally
npm install -g defuddle

# Verify
defuddle --version

Note: Synapse finds defuddle automatically via nvm paths even if it's not on your shell's PATH. If wiki_fetch_url reports defuddle not found, ensure it's installed in an nvm-managed Node version.

Obsidian Vault Setup

  1. Create a new vault in Obsidian (or clone your wiki repo)

  2. Install the Git community plugin (Settings β†’ Community Plugins β†’ Browse β†’ "Git")

  3. Configure Git plugin with your GitHub credentials

  4. The vault structure (raw/, wiki/, Clippings/, AGENTS.md) is created automatically by Synapse on first run

Installation

cd /path/to/your/workspace
git clone <repository-url> project-synapse-mcp
cd project-synapse-mcp
uv venv --python 3.12 --seed
source .venv/bin/activate
uv add -e .
uv run python -m spacy download en_core_web_sm
cp .env.example .env  # edit with your Neo4j password and vault path

Configuration

Edit .env:

NEO4J_URI=bolt://localhost:7687
NEO4J_USER=neo4j
NEO4J_PASSWORD=your_password
NEO4J_DATABASE=neo4j

# Embedding β€” local only, no paid APIs
EMBEDDING_PROVIDER=sentence-transformers  # or "ollama"
EMBEDDING_MODEL=sentence-transformers/all-mpnet-base-v2
EMBEDDING_DIMENSION=768

# Extraction β€” Montague (default) or "llm" (hybrid)
EXTRACTION_PROVIDER=montague
OLLAMA_EXTRACTION_MODEL=gemma2:9b
OLLAMA_TIMEOUT=120

# Wiki vault
WIKI_VAULT_PATH=/path/to/your/obsidian-vault
WIKI_GITHUB_REPO=https://github.com/user/wiki-repo

Claude Desktop / MCP Integration

Add to your MCP config:

{
  "mcpServers": {
    "project-synapse": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/project-synapse-mcp",
        "run",
        "python",
        "-m",
        "synapse_mcp.server"
      ]
    }
  }
}

MCP Tools

Knowledge Graph

Tool

Description

ingest_text

Process text through semantic pipeline β†’ Neo4j

query_knowledge

Vector semantic search with insight-first results

explore_connections

Graph traversal for hidden relationships

generate_insights

Autonomous Zettelkatten pattern detection

analyze_semantic_structure

Montague Grammar semantic analysis

Wiki (LLM-WIKI)

Tool

Description

wiki_fetch_url

Fetch URL β†’ defuddle clean β†’ ingest β†’ archive to Clippings/

wiki_ingest_raw

Ingest file from raw/ β†’ Neo4j + auto-move to Clippings/

wiki_write_page

Create/update wiki page with frontmatter (updates index write-through)

wiki_read_page

Read a wiki page by path supporting mode (meta, excerpt, full)

wiki_search

Keyword search across wiki pages returning excerpts and snippets

wiki_list_pages

List pages in a subdirectory with paginated limit, offset, and tag filters

wiki_update_index

Rebuild the wiki index (index.md)

wiki_sync_index

Manually sync/refresh the DuckDB page index database from disk

wiki_lint

Health check: orphans, broken links, missing/invalid frontmatter (runs via SQL)

Wiki Vault Structure

LLM-WIKI/
β”œβ”€β”€ AGENTS.md           # Agent schema doc β€” conventions and workflows
β”œβ”€β”€ raw/                # INBOX ONLY β€” unprocessed files; empty after each session
β”œβ”€β”€ raw-inbox.base      # Obsidian Base view of pending raw/ queue
β”œβ”€β”€ Clippings/          # Permanent archive β€” all processed sources land here
β”œβ”€β”€ wiki/
β”‚   β”œβ”€β”€ index.md        # Auto-generated page catalogue
β”‚   β”œβ”€β”€ log.md          # Append-only activity log
β”‚   β”œβ”€β”€ entities/       # People, tools, projects
β”‚   β”œβ”€β”€ concepts/       # Ideas, theories, patterns
β”‚   └── sources/        # Summaries of ingested sources

Content Lifecycle

You clip/save β†’ raw/          # your inbox
     or
Agent fetches β†’ wiki_fetch_url # web research
                    β”‚
              [defuddle clean]
                    β”‚
              [semantic pipeline] β†’ Neo4j
                    β”‚
              wiki_write_page β†’ wiki/sources/
                    β”‚
              auto-move β†’ Clippings/   # permanent archive

raw/ is always empty after a session. Clippings/ is the permanent record of everything that's been processed. Source pages in wiki/sources/ reference the original URL, not the file path.

Workflow

  1. Web research: wiki_fetch_url(url) β†’ fetches, cleans, ingests, archives in one call

  2. Manual clip: Drop into raw/, call wiki_ingest_raw(filename) β†’ auto-archives after ingest

  3. Query: query_knowledge (graph) or wiki_search (files) β†’ synthesize answer

  4. Lint: wiki_lint β†’ fix orphans, broken links, stale claims

  5. Rollback: Git handles version control via Obsidian Git plugin

Theoretical Foundation

  • Montague Grammar: Formal compositional semantics for meaning extraction

  • Zettelkasten Method: Atomic linked notes with emergent structure

  • Graph Theory: Community detection, centrality, path analysis

  • Karpathy LLM-WIKI: Persistent knowledge compilation vs stateless RAG

  • Vannevar Bush's Memex: Private associative knowledge with maintained trails

License

MIT β€” see LICENSE.


Project Synapse: From reactive RAG to persistent, compounding knowledge.

Available Tools

23 tools
analyze_semantic_structureB

Analyze the semantic structure of text using Montague Grammar parsing.

This tool provides insight into the formal semantic analysis capabilities and shows the logical form translations.

Args: text: Text to analyze semantically include_logical_form: Whether to include the formal logical representation

Returns: Semantic analysis with entities, relations, and optional logical forms

Inputs: text (required), include_logical_form. Returns: result

debug_testB

Simple test tool to check if MCP server is working.

Inputs: no named parameters. Returns: result

explore_connectionsB

Explore connections and relationships around a specific entity in the knowledge graph.

This tool implements the graph traversal capabilities for discovering non-obvious connections and patterns.

Args: entity: Entity name to explore from depth: How many relationship hops to explore (1-5) connection_types: Specific relationship types to follow

Returns: Visual representation of connections and discovered patterns

Inputs: entity (required), depth, connection_types. Returns: result

generate_insightsC

Trigger autonomous insight generation using the Zettelkasten engine.

This tool activates the autonomous synthesis engine to identify patterns and generate novel insights from the existing knowledge graph.

Args: topic: Optional topic to focus insight generation on confidence_threshold: Minimum confidence level for insights (0.0-1.0)

Returns: Generated insights with confidence scores and evidence trails

Inputs: topic, confidence_threshold. Returns: result

ingest_textB

Ingest and process text into the knowledge graph using semantic analysis.

This tool performs the core knowledge synthesis pipeline:

  1. Semantic parsing using Montague Grammar

  2. Entity extraction and relationship identification

  3. Storage in the Neo4j knowledge graph

  4. Automatic insight generation triggers

Args: text: Raw text to process and analyze source: Source identifier for provenance tracking metadata: Additional metadata about the text

Returns: Processing summary with entities and relationships extracted

Inputs: text (required), source, metadata. Returns: result

query_knowledgeA

Query the knowledge graph for facts and insights using natural language.

This tool provides the conversational interface to the knowledge base, prioritizing synthesized insights over raw facts.

Args: query: Natural language query include_insights: Whether to include AI-generated insights max_results: Maximum number of results to return

Returns: Query results with facts, insights, and reasoning trails

Inputs: query (required), max_results, include_insights. Returns: result

synapse_causal_windowA

Find candidate causes by temporal correlation.

Surfaces facts whose valid_from falls in the window [before - within_days, before] and that share at least one entity with facts about effect_entity.

This is exactly the "track everything you ate to find what caused the headaches" pattern β€” you record symptom onset, you record meals and medications, then this tool surfaces co-occurring events as candidates. The tool returns correlation; the human (or a downstream reasoning step) decides what caused what.

Args: effect_entity: The thing whose causes you're hunting (e.g. "headache", "rash", "build failure"). before: ISO date/datetime β€” when the effect was observed. within_days: How far back to search. Default 30.

Returns: Ranked list of candidate cause-effect pairings with day deltas.

Inputs: before (required), effect_entity (required), within_days. Returns: result

synapse_invalidateA

Mark a previously-recorded fact as no longer true.

Sets valid_to rather than deleting β€” the historical record stays intact, but the fact is no longer "currently true" for default queries.

Args: subject/predicate/object: The triple to invalidate. ended: ISO date/datetime when the fact stopped being true. Defaults to now if omitted.

Returns: Number of facts affected.

Inputs: object (required), subject (required), predicate (required), ended. Returns: result

synapse_memory_statsA

Quick stats: how many temporal facts are stored, time span covered.

Inputs: no named parameters. Returns: result

synapse_recallA

Look up time-stamped facts about an entity.

Args: entity: Name to look up. Matches subject, object, or both depending on direction. as_of: ISO date/datetime β€” if given, only facts valid at this point in time are returned. Omit for "currently true" facts. direction: "outgoing" (entity is the subject), "incoming" (entity is the object), or "both" (default).

Returns: Newline-separated list of facts with timestamps. Empty if none found.

Inputs: entity (required), as_of, direction. Returns: result

synapse_rememberA

Record a time-stamped fact in Synapse's episodic memory.

Use this whenever something is worth remembering across sessions: decisions, observations, "Ty said X on date Y", health/diet/symptom log entries, project milestones.

Args: subject: Who or what the fact is about. Free-form name. predicate: The relationship verb. snake_case preferred (e.g. "started_taking", "moved_to", "decided_to_use"). object: The other side of the relation. valid_from: ISO date or datetime when the fact became true. If omitted, "now" is used. Bare dates β†’ midnight UTC. valid_to: ISO date or datetime when the fact stopped being true. Omit for still-current facts. confidence: 0–1. Default 1.0 for explicit user statements; lower when the agent is inferring. source: Where this fact came from. Defaults to "agent:claude" for things Claude is recording. Use "user" or a filename for facts from explicit user statements or document ingestion. note: Free-form context. Stored in metadata for later recall.

Returns: The fact id (stable content hash β€” safe to call twice).

Inputs: object (required), subject (required), predicate (required), note, +4 more. Returns: result

synapse_timelineA

Chronological view of remembered facts.

Args: entity: Scope to one entity, or None for the global timeline. limit: Max number of rows. Default 50.

Returns: Time-ordered fact list, oldest first.

Inputs: limit, entity. Returns: result

wiki_cluster_pagesA

Cluster wiki pages by semantic similarity using GAAC (TF-IDF).

Identifies:

  • Natural topic clusters β€” pages that belong together

  • Missing links β€” same-cluster pages with no wikilink between them

  • Merge candidates β€” pages so similar they may be redundant (sim > 0.7)

Args: n_clusters: Number of clusters (auto = sqrt of page count if omitted).

Inputs: n_clusters. Returns: result

wiki_fetch_urlA

Fetch a URL with defuddle (clean markdown extraction), save to raw/, ingest into the knowledge graph, and archive to Clippings/.

Use this when researching the web β€” it strips navigation and clutter, leaving only the article content. Much cleaner than raw web_fetch.

Args: url: The URL to fetch and process. ingest: If True (default), immediately ingest into Neo4j after saving. Set False to save to raw/ only for manual review first.

Inputs: url (required), ingest. Returns: result

wiki_hits_analysisA

Compute HITS hub and authority scores on the wiki wikilink graph.

Authorities = pages cited by many others β€” load-bearing knowledge nodes. Hubs = pages that link to many good authorities β€” navigation layers.

Use to identify which pages need deepening (high authority) and which need comprehensive link coverage (high hub).

Inputs: no named parameters. Returns: result

wiki_ingest_rawA

Read a raw source file and ingest it into both the knowledge graph and wiki.

Reads from raw/, runs it through the Synapse semantic pipeline, stores in Neo4j, and creates a summary page in wiki/sources/.

Args: filename: Filename inside the raw/ directory.

Inputs: filename (required). Returns: result

wiki_lintA

Run a health check on the wiki vault.

Detects orphan pages, broken wikilinks, and missing frontmatter.

Inputs: no named parameters. Returns: result

wiki_list_pagesB

List all markdown pages in the wiki vault with pagination.

Args: subdir: Subdirectory to list ('wiki' or 'raw'). limit: Maximum pages to return (default 50, max 200). offset: Offset for pagination. tag: Filter by tag.

Inputs: tag, limit, offset, subdir. Returns: result

wiki_read_pageB

Read a wiki page.

Args: path: Relative path from vault root (e.g. 'wiki/concepts/rag.md'). mode: What to return: - "meta": metadata/frontmatter only (from index, 0 file reads) - "excerpt": metadata + first 500 characters of page body (from index) - "full": complete page content (reads file from disk)

Inputs: path (required), mode. Returns: result

wiki_sync_indexA

Refresh the wiki index database from disk.

Runs automatically on server start and after write operations. Call manually if files were changed outside Synapse (e.g., Obsidian edits, git pull).

Inputs: no named parameters. Returns: result

wiki_update_indexB

Rebuild the wiki index from all wiki pages.

Args: deep: If True, performs a disk-level verification of all indexed files.

Inputs: deep. Returns: result

wiki_write_pageB

Write or update a wiki page with frontmatter.

Args: path: Relative path (e.g. 'wiki/entities/neo4j.md'). body: Markdown body content. summary: One-line summary for the index. tags: Comma-separated tags.

Inputs: body (required), path (required), tags, summary. Returns: result

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    B3.4/5.0

    Scored across 23 tools

    Disambiguation3/5

    The tools are grouped into wiki, knowledge graph, and memory domains, but several have blurry boundaries: wiki_sync_index vs wiki_update_index, synapse_recall vs synapse_timeline, and ingest_text vs wiki_ingest_raw could cause misselection. Descriptions help clarify most overlaps, but not all.

    Naming Consistency4/5

    Most tools follow verb_noun snake_case with clear wiki_ and synapse_ prefixes, but the core knowledge graph tools (ingest_text, query_knowledge, explore_connections) lack a shared prefix and debug_test breaks the pattern. The naming is mostly consistent but not fully systematic.

    Tool Count3/5

    23 tools is in the heavy range and requires a lot of surface area to navigate, though each tool has a plausible role in the wiki/knowledge-graph/memory system. It is borderline: not overwhelming, but more than is clearly necessary.

    Completeness4/5

    The set covers wiki read/write/indexing, knowledge graph ingestion and querying, and episodic memory lifecycle management including invalidation and causal analysis. Minor gaps exist (no wiki page deletion, no explicit knowledge graph entity update/delete), but core workflows are well covered.

    Maintenance

    ActivityInactive
    ResponsivenessLimited data

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      D
      maintenance
      An MCP server that transforms markdown notes into an AI-powered knowledge graph. It enables LLM clients to explore, analyze, and diagnose knowledge graphs through tools for node explanation, path finding, causal chain analysis, and wiki health reporting.
      5
      MIT
    • F
      license
      Not graded
      quality
      D
      maintenance
      An MCP server that implements a heavily typed knowledge graph memory system with AI-powered entity and relation extraction, enabling structured knowledge storage and retrieval from unstructured text using predefined or custom ontologies.
      9
      -
    • A
      license
      Not graded
      quality
      D
      maintenance
      An MCP server that compiles any text into a verifiable, graph-based knowledge base using deterministic chunking and parallel extraction of epistemology primitives.
      MIT
    • A
      license
      A
      quality
      B
      maintenance
      An MCP server that turns any compatible AI agent into a Zettelkasten partner for creating atomic notes, forming semantic links, detecting clusters, and synthesizing insights from your existing knowledge.
      19
      2
      MIT