MCP.so
Sign In

ForkMind

@Medhovarsh

About ForkMind

🧠 Local-first LLM state branching & debugging. Capture, visualize, branch, and regression-test LLM calls as a DAG. Free & local via Ollama, any OpenAI-compatible API, MCP for agents. Zero config.

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "forkmind": {
      "command": "npx",
      "args": [
        "-y",
        "forkmind",
        "mcp"
      ]
    }
  }
}

Tools

15

List the most recent captured LLM turns (newest first), compact. Use to recall what was just tried.

Fetch one node by id with full request + response payloads.

Return the full conversation path from the root to the given node β€” the exact context that produced it. Use to understand how a state was reached.

List the alternative continuations (branches) that fork from a node. Use to compare what different prompts/params produced from the same point.

Case-insensitive substring search across all captured requests/responses. Use to find prior attempts mentioning a term, error, or tool.

Summary of the conversation tree: totals, roots, leaves, providers.

Offload conversation context into an immutable, encrypted DAG on disk. Pass the items to archive plus a short digest YOU write (the retrieval key you keep in your window; omit for private capsules). After the id comes back, verify with forkmind_context_digest, THEN drop the material from your working context β€” never before.

Compact list of saved capsules (title, digest, size, age). Optional substring filter over title + digest.

Cheap probe: full digest and DAG structure for one capsule, no decryption of content. Use before restoring to decide whether you need the whole capsule or one segment. Also serves as the durability check after a save, before you compact your window.

Decrypt and return capsule content (integrity-verified first). Pass segmentIds for a partial restore β€” pull back only what you need.

IRREVERSIBLY destroy a capsule: its encryption key is shredded and the id tombstoned. You must echo the capsule id in `confirm`. Only call when the user explicitly asks to forget.

Aggregate stats across all capsules: count, bytes, estimated tokens freed from the window, how many have a digest, forgotten count, and replica health.

Export a capsule as a passphrase-encrypted, self-contained bundle that can move to another project or machine β€” it carries its own key material and does not depend on this project's local master key. The passphrase is never stored in the bundle; the caller must keep it to import later. Returns the bundle JSON β€” write it to a file.

Import a bundle produced by forkmind_context_export. Every segment is independently re-verified (id, hash, parent graph, acyclicity) before anything is written β€” the bundle is never trusted blindly. Re-wraps the capsule under this project's local master key.

Status of redundant capsule storage: each replica target with reachability and coverage. Optionally sync (push all capsules + propagate tombstones) before reporting.

Overview

What is ForkMind?

ForkMind is a local-first LLM state branching and debugging tool that treats AI context windows like a Git repository. It captures every LLM call into a local .forkmind/ directory, visualizes the conversation as a Directed Acyclic Graph (DAG), and lets you branch alternative prompts or model parameters from any point in historyβ€”all on your machine with no cloud or account. It works with any OpenAI-compatible API, defaulting to free open-source models via Ollama, and also supports Anthropic and hosted free tiers (Groq, OpenRouter, Together, vLLM, LM Studio).

How to use ForkMind?

Install ForkMind via npm (npx forkmind init, npx forkmind start) or globally (npm install -g forkmind). Quick start: install Ollama, pull a model, run the CLI to start a proxy on http://localhost:4500, then point your code at that proxy. ForkMind provides a drop-in SDK for Node.js (ForkMindOpenAI), adapters for LangChain.js and Vercel AI SDK, and a Claude Code plugin. Any language can capture calls by setting the client’s base URL to http://localhost:4500/v1 and passing the x-forkmind-upstream and x-forkmind-parent headers.

Key features of ForkMind

  • Local-first LLM state branching and debugging.
  • Conversation visualized as a Directed Acyclic Graph.
  • Branch from any point with alternative prompts or models.
  • Compatible with any OpenAI-compatible API or Anthropic.
  • MCP server for AI agents to query their own history.
  • Plain JSON storage, no database or telemetry.

Use cases of ForkMind

Comments

More AI & Agents MCP servers