Skip to content
MCP server

Inwardis

Let agents read, search and edit a self-hosted architecture model over MCP

Pricing
Free trial
Platforms
self-hosted, API, vscode
Runs
Remotely
Version
1.0.22
Listing updated
2 days ago

About Inwardis

Inwardis is an MCP server for developer tools. It runs as a hosted remote server, so there is nothing to install locally.

Inwardis models a system as layers: context, containers, components, decisions and data, each level summarising the one below. Its MCP server ships inside the Inwardis server you run with Docker, so the model stays on your own infrastructure.

Agents call learn first, which explains how the model is organised, then descend only where a task needs it. They can trace relations in both directions to answer what depends on something, apply atomic batches of changes, and call render_view to get an SVG with a summary of what was drawn so they can spot clipped labels. Changes appear in the activity feed under the API key's name.

Access uses a personal bearer key that is scoped, revocable and rate limited per key. A fresh install has no keys, and an administrator can turn the endpoint off. The page lists 61 tools in groups for reading, building, views, review, history and code, and import or export.

Tools this server provides (8)

Taken from the project documentation; the running server may differ.

learn
Teaches the agent how the model is organised and how to read it level by level
project_brief
Summarise a project
read_project_tree
Read a project's element tree
search_elements
Search elements in the model
trace
Follow relations from an element
apply_changes
Apply a batch of model changes
render_view
Render a view as SVG with a summary of what it drew
add_comment
Add a comment on the model

What it does

Layered reading
Agents read the top level and descend only where the task needs detail.
Dependency tracing
Relations and cross-project references are followed both ways.
Atomic batch changes
apply_changes creates or edits projects, elements and relations in one batch.
Self-checking diagrams
render_view returns SVG plus a summary of what was actually drawn.
Scoped API keys
Keys are revocable, rate limited, and limited to what their owner can see.
Activity trail
Every change shows in the feed under the key's name, with optional session notes.

Good for

  • Let a coding agent read architecture before it writes code
  • Ask what depends on a component before changing it
  • Have an agent draft or correct a C4-style model
  • Link code references and git repositories to model elements

Pricing

Free trial

30-day trial; prices not published

Prices change. Check inwardis.com before you buy.

Inwardis FAQ

Where does the model live?

On your own server; the MCP server is part of the Inwardis installation.

How do clients authenticate?

With an Authorization bearer key starting ink_, created under API Keys in the app.

Which clients are documented?

Claude Code, Cursor and VS Code, plus any client that speaks MCP over HTTP with a custom header.

Is there a trial?

A 30-day trial includes every feature, the MCP server among them.

Is Inwardis free?

Inwardis is a paid product with a free trial.

What are the best alternatives to Inwardis?

The closest alternatives to Inwardis are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to Inwardis

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools