malimgraph 0.1.4

Transform PDF documents into structured knowledge graphs with citation provenance

MalimGraph

███╗   ███╗ █████╗ ██╗     ██╗███╗   ███╗ ██████╗ ██████╗  █████╗ ██████╗ ██╗  ██╗
████╗ ████║██╔══██╗██║     ██║████╗ ████║██╔════╝ ██╔══██╗██╔══██╗██╔══██╗██║  ██║
██╔████╔██║███████║██║     ██║██╔████╔██║██║  ███╗██████╔╝███████║██████╔╝███████║
██║╚██╔╝██║██╔══██║██║     ██║██║╚██╔╝██║██║   ██║██╔══██╗██╔══██║██╔═══╝ ██╔══██║
██║ ╚═╝ ██║██║  ██║███████╗██║██║ ╚═╝ ██║╚██████╔╝██║  ██║██║  ██║██║     ██║  ██║
╚═╝     ╚═╝╚═╝  ╚═╝╚══════╝╚═╝╚═╝     ╚═╝ ╚═════╝ ╚═╝  ╚═╝╚═╝  ╚═╝╚═╝     ╚═╝  ╚═╝

From documents to knowledge graphs.

Agentic knowledge graph plugin for Claude Code, Claude Desktop, and Codex. Extract entities, build graphs, chunk for RAG, render HTML, and load into Neo4j or pgvector — all orchestrated by Claude using its own intelligence. No ANTHROPIC_API_KEY required.

---

Install

pip install malimgraph
claude mcp add malimgraph -- malimgraph-plugin

Then just ask Claude naturally:

"Extract a knowledge graph from report.pdf" "Chunk annual_report.pdf for RAG and store in pgvector" "Full pipeline on this document"

---

How It Works

You: "Extract a knowledge graph from report.pdf"
        │
        ▼
Claude calls  read_pdf("report.pdf")
        │     ← returns page text + rule entities (dates, amounts, emails…)
        │
        ▼
Claude analyzes text   ← uses YOUR Claude subscription, no extra API key
        │     identifies: Organizations, People, Regulations, Events…
        │     maps: relationships with verbatim source_text evidence
        │
        ▼
Claude calls  save_knowledge_graph(entities, relationships, output_format="all")
        │     ← builds KnowledgeGraph, saves files
        ▼
  ./output/
    ├── knowledge_graph.json    ← full graph with provenance
    ├── knowledge_graph.cypher  ← Neo4j import
    └── knowledge_graph.sql     ← Apache AGE import

---

Skill Triggers

Say these phrases to activate built-in workflows:

Phrase	Workflow	Tools
"knowledge graph" / "extract entities"	$pdf-to-graph	read_pdf → save_knowledge_graph
"chunk for RAG" / "vector search" / "pgvector"	$pdf-to-rag	chunk_document → embed_and_store_chunks
"full pipeline" / "extract and embed"	Full pipeline	All tools in sequence
"load into Neo4j" / "Cypher query"	$graph-query	manage_graph_db
"render HTML" / "browsable document"	$document-html	render_document_html

---

Available Tools

Tool	Description
read_pdf	Parse PDF → page text + rule entities. First step of any KG workflow.
save_knowledge_graph	Accept Claude-extracted entities/relationships → save .json/.cypher/.sql
chunk_document	Token-aware overlapping chunks with heading context for RAG
render_document_html	Structured HTML with page anchors, entity annotations, TOC + search
manage_graph_db	Load, query, and manage graphs in Neo4j or Apache AGE
embed_and_store_chunks	Embed chunks into PostgreSQL pgvector (OpenAI / Voyage / local)
list_workflows	List all available workflows, triggers, and tool sequences

---

Runtimes

Runtime	Install
Claude Code	claude mcp add malimgraph -- malimgraph-plugin
Claude Desktop	See config below
Codex / OpenAI Agents	See AGENTS.md for function schemas
Any MCP runtime	{"command": "malimgraph-plugin"}

Claude Desktop (claude_desktop_config.json):

{
  "mcpServers": {
    "malimgraph": {
      "command": "malimgraph-plugin"
    }
  }
}

---

CLI (standalone, requires ANTHROPIC_API_KEY)

export ANTHROPIC_API_KEY=sk-ant-...

# Full pipeline
malimgraph extract --input report.pdf --output ./output/ --format all
malimgraph chunk --input report.pdf --output ./chunks/
malimgraph render --input report.pdf --output document.html

# pgvector
export PGVECTOR_URI="postgresql://user:pass@localhost:5432/mydb"
export OPENAI_API_KEY=sk-...
malimgraph vector load --input ./chunks/chunks.json

# Graph database
malimgraph db load --input ./output/knowledge_graph.json \
  --target neo4j --uri bolt://localhost:7687 --user neo4j --password secret
malimgraph db query --target neo4j --uri bolt://localhost:7687 \
  --query "MATCH (n:Organization) RETURN n.label, n.source_pages LIMIT 10"

---

Installation Options

pip install malimgraph                    # core
pip install "malimgraph[neo4j]"           # + Neo4j driver
pip install "malimgraph[pgvector,openai]" # + pgvector + OpenAI embeddings
pip install "malimgraph[pgvector,voyage]" # + pgvector + Voyage AI
pip install "malimgraph[pgvector,local]"  # + pgvector + local CPU embeddings
pip install "malimgraph[all]"             # everything

---

Output Schema

Every entity and relationship carries full citation provenance:

Field	Description
id	Stable hash: e_ + MD5(type:label)[:8]
label	Canonical entity name
type	Organization / Person / Location / Regulation / …
source_text	Verbatim quote from the document
source_pages	PDF page numbers
confidence	high / medium / low
extraction_method	rule / llm / hybrid
citations[]	All supporting quotes with page refs

---

pgvector Embedding Providers

Provider	Default model	Dimension	Requires
openai	text-embedding-3-small	1536-d	OPENAI_API_KEY
voyage	voyage-3-large	1024-d	VOYAGE_API_KEY
local	all-MiniLM-L6-v2	384-d	none (CPU)

---

Database Setup

# Neo4j
docker run -p 7474:7474 -p 7687:7687 \
  -e NEO4J_AUTH=neo4j/yourpassword neo4j:latest

# Apache AGE
docker run -p 5432:5432 -e POSTGRES_PASSWORD=secret apache/age:latest

# pgvector
docker run -p 5432:5432 -e POSTGRES_PASSWORD=secret pgvector/pgvector:pg17

See docs/database-setup.md for full guides.

---

Contributing

git clone https://github.com/malim-ai-labs/malim-graph-plugin
pip install -e ".[dev]"
make test
make lint

---

Credits

Built by Malim AI Labs — AI-powered knowledge infrastructure for Southeast Asia.

Malim AI Labs Social Enterprise (003827047-U) · Kuala Lumpur, Malaysia

---

License

MIT — see LICENSE