endiagram
The endiagram server performs deterministic, graph-theoretic structural analysis of systems modeled in EN (Entity-Node) syntax — no AI inside the computation.
analyze_system— Compute structural facts (shape, stages, bridges, cycles, actors, locations); optionally check invariants and detect antipatterns (unguarded-sinks, single/multi-cut-paths), liveness, boundedness, siphons, and traps.render— Generate publication-quality SVG diagrams with configurable light/dark themes.detail— Deep structural analysis: concurrency, flow landmarks, resilience metrics, dominator trees, and min-cuts.distance— Find the shortest path between two nodes, annotated with subsystem crossings.trace— Follow directed flow from node A to node B with role/subsystem annotations and optional defense-node highlighting.diff— Structurally compare two systems, identifying topology, role, and subsystem changes.extract— Extract a named subsystem as standalone EN source code.impact— Blast-radius analysis: remove a node and see what disconnects.evolve— Dry-run architectural changes by applying a patch and viewing the structural delta.between— Compute betweenness centrality for a specific node (fraction of shortest paths through it).categorize— Auto-discover subsystem boundaries from the dependency structure.compose— Merge two EN graphs into one with explicit entity linking.
Allows rendering complex system dependency graphs and structural analysis results into publication-quality SVG images.
@endiagram/mcp
MCP server for EN Diagram — deterministic structural analysis powered by graph theory. Every result is backed by a named mathematical theorem. No AI inside the computation.
Installation
Run directly:
npx @endiagram/mcpOr install globally:
npm install -g @endiagram/mcpRelated MCP server: systems-thinking-mcp
Connect
Claude Code
claude mcp add endiagram npx @endiagram/mcpClaude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"endiagram": {
"command": "npx",
"args": ["@endiagram/mcp"]
}
}
}Cursor
Add to .cursor/mcp.json in your project root:
{
"mcpServers": {
"endiagram": {
"command": "npx",
"args": ["@endiagram/mcp"]
}
}
}HTTP (zero install)
Any MCP client that supports HTTP transport:
https://api.endiagram.com/mcpSmithery
smithery mcp add dushyant30suthar/endiagramEnvironment Variables
Variable | Default | Description |
|
| API endpoint for the EN Diagram service |
Tools
Six questions about any system, plus a render tool. Every tool takes source (EN code or .en/.txt file path). Tool names are shorthand, not specs — read each description before calling; compose and equivalent are mode-based, not general analyzers.
Tool | What it answers | Levers |
| What is this system? Shape, stages, bridges, cycles, critical path, dominator tree, min-cuts, subsystems, actors, locations. |
|
| What's always true? Conservation laws, T-invariants (sustainable cycles), depletable sets, deficiency, reversibility. |
|
| Can it deadlock? Can entities overflow? Siphons, traps, unbounded cycles, structural liveness and boundedness. | — |
| Can X reach Y? Path, distance, boundary crossings. |
|
| Are two systems the same, or what changes if I change this one? | Compare mode ( |
| How do parts combine (merge) or how does a part stand alone (extract)? | Merge: |
| SVG or PNG diagram. Only call when the user asks to visualize. | Themes: |
EN Syntax
One statement per line:
actor do: action needs: input1, input2 yields: output1, output2Shared names between yields and needs create connections automatically:
customer do: place order needs: menu yields: order
kitchen do: prepare food needs: order yields: meal
waiter do: deliver needs: meal yields: served customerModeling
Same name = same thing. Put all required inputs in one needs: list (AND). Give two actions the same yield-name to offer alternatives (OR). Sequence = one action consuming another's yield. Re-yield stateful resources to keep them persistent; one-shot data stays consumed. Model at the real atomic granularity — split only when the pieces could be accessed independently.
Learn more at endiagram.com.
Telemetry
@endiagram/mcp generates a random install ID on first run, stored at
~/.endiagram/install-id (mode 0600). It is sent with every request as
the X-Endiagram-Install-Id HTTP header so we can correlate requests
from the same install for debugging issues that the per-IP signal alone
cannot track (mobile networks, VPNs, CGNAT all collapse or churn IPs).
No source code, no file paths, no environment variables, and no PII are sent. The install ID is a random opaque UUIDv4 generated locally.
A first-run notice prints to stderr (never stdout — stdout is the MCP JSON-RPC channel) with the disclosure and the opt-out instructions. The notice fires once per install and never again.
Opting out
Any of these three methods disables the install ID:
Set
ENDIAGRAM_TELEMETRY=offas an environment variable (also accepts0,false,no).Create a file at
~/.endiagram/telemetrycontaining the wordoff.Delete
~/.endiagram/install-id. (A new one is generated on next run unless option 1 or 2 is also set.)
When any of these is active, the X-Endiagram-Install-Id header is not
sent at all — the server falls back to its per-IP HMAC cid for
correlation, which works fine for short-term per-session tracing.
Full privacy policy: endiagram.com/privacy
License
MIT
Available Tools
12 toolsanalyze_systemCInspect
Structural signal. You describe the system, the tool computes structural facts. All computation is deterministic — no AI inside. EN syntax: subject do: action needs: inputs yields: outputs.
Inputs: source (required), invariants, detect_antipatterns.
betweenBInspect
Betweenness centrality for a node — what fraction of all shortest paths flow through it.
Inputs: source (required), node (required).
categorizeCInspect
Auto-discover subsystem boundaries from dependency structure.
Inputs: source (required).
composeCInspect
Merge two EN graphs into one with entity linking.
Inputs: source_a (required), source_b (required), links (required).
detailCInspect
Deep structural analysis — concurrency, flow landmarks, resilience, dominator tree, min-cuts.
Inputs: source (required).
diffCInspect
Structural diff between two systems — topology, role, and subsystem changes.
Inputs: source_a (required), source_b (required).
distanceCInspect
Shortest path between two nodes with subsystem crossing annotations.
Inputs: source (required), from (required), to (required).
evolveCInspect
Dry-run architectural changes — apply a patch and see the structural delta.
Inputs: source (required), patch (required).
extractBInspect
Extract a named subsystem as standalone EN source code.
Inputs: source (required), subsystem (required).
impactCInspect
Blast radius — remove a node and see what disconnects.
Inputs: source (required), node (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 12 tools
Most tools target distinct graph operations, but trace, distance, and between all revolve around paths and centrality, while analyze_system, detail, and categorize overlap in structural analysis. Descriptions help but an agent could reasonably pick the wrong tool for some analytical tasks.
Tool names are mostly single lowercase words, but they mix verbs, nouns, and prepositions (between, distance, impact, detail), and analyze_system breaks the pattern with snake_case. There is no consistent verb_noun or predictable naming convention.
Twelve tools is well-scoped for a dependency-graph analysis server. Each tool has a recognizable responsibility within analysis, rendering, transformation, and composition, without feeling padded or thin.
The toolkit covers traversal, centrality, structural diffing, subsystem discovery, deep analysis, rendering, merging, and dry-run evolution. Minor gaps like explicit node creation/deletion or cycle detection exist, but the server's analytical focus is well covered.
Maintenance
Related MCP Connectors
AI Agent with Architectural Memory. Impact analysis (free), tests and code from the graph (pro).
30+ deterministic dev and delivery tools: linting, metrics, diagrams, model advice, AI-text checks.
Code intelligence platform for AI agents. 20 tools for architecture, security & impact analysis.
Deterministic reasoning stack for AI agents: simulate, decide & compute, plus cross-domain tools.
Related MCP Servers
- -licenseNot gradedqualityNot gradedmaintenanceEnables AI agents to traverse SysML v2 model graphs, query requirements, and perform impact analysis for model-based systems engineering. It allows agents to interact with plain-text models to automate documentation and refine system architectures.-
- FlicenseAqualityDmaintenanceProvides systems thinking models as composable analysis lenses for architecture, infrastructure, DevOps, incident analysis, and technical decision-making.51-
- AlicenseBqualityCmaintenanceWhole-repo code graph for AI agents: impact analysis, dead code, cycles, semantic code search — 14 tools via hosted MCP endpoint.15MIT

Gridlaceofficial
AlicenseNot gradedqualityBmaintenanceEnables deterministic dependency reasoning over declared graphs, allowing users and AI agents to validate acyclicity, resolve targets, identify runnable parallel layers, assess impact, and explain blockers.Apache 2.0
Appeared in Searches
- Tools for visualizing NetworkX graphs without an API key
- Free Graph and Network Visualization Tools (No API Key Required)
- Interactive Network Graph Visualization Tools (No Account Required)
- Free Network Graph Visualization Tools (No Account Required)
- Advanced Engineering Fundamentals: Structural, Materials, Dynamics, and Thermodynamics