Codebase intelligence for AI teams

Your team plans it.
Your agents build it.
Everyone sees.

Mycelium connects your team's planning to your agents' execution — a shared canvas they both read from, a codebase graph they both navigate, and a history that captures everything.

$ npx @kopikocappu/mycelium init
Claude Code · Cursor · GitHub Copilot · Any MCP agent
↓ scroll
01 — Plan

Your team draws the map.

The roadmap canvas is where humans decide what gets built — features, dependencies, priorities, critical path. Your agents can read it all through the same MCP server. They know what's planned before you assign anything.

02 — Execute

Your agents follow it precisely.

Before touching a file, your agent calls /preflight and gets back exactly which files to read. When it's done, it marks the feature complete on the canvas. The loop closes automatically.

03 — See

Everyone sees what happened.

Every file touched, every session logged, every change attributed — to a human, Claude Code, Cursor, or any agent. The canvas updates. The history fills in. Nothing your agents do is invisible.

The planning layer your agents actually understand.

Draw your features. Set dependencies. Find your critical path. Your agents read the exact same canvas through the MCP server — they know what's ready to build, what's blocked, and what to work on next. When they're done, they mark it complete. You see it instantly.

This isn't a project management tool bolted on. It's the shared source of truth between you and your AI agents.

Graph
⬢ Roadmap
History
tier 0 tier 1 tier 2 Auth DB Schema UI Kit API Layer Stripe Dashboard Checkout Analytics ● CRITICAL PATH Complete Active Planned Blocked + feature ● Path ⌘K ↵ Undo
Core REST layer for all client-facing features. Checkout, Dashboard, and Analytics all depend on this being complete.
WHAT'S BLOCKING
✓ All deps complete — ready
Critical path
● Live
Longest incomplete dep chain — shown in amber. Shorten it and everything ships faster.
Agent-readable
MCP
Agents read the canvas via GET /roadmap and mark features done via PATCH.
Supertree / Subtree
Isolate any feature to see only its prerequisites or its full set of dependents.
Snapshots
Save named versions of the roadmap state. Restore any snapshot with full undo support.

Everything the loop needs to work.

For your agents

Natural language preflight

Describe any task in plain English. Get back exactly the files that matter, scored by relevance, with AI-written descriptions. Four files instead of forty, every time.

Semantic search

Find files by concept, not keyword. "Authentication flow" surfaces the right files even when none of them have the word "auth" in their name. The graph understands what code means.

Cross-reference analysis

Before modifying a function, ask what calls it. Get the full blast radius — every file and function that depends on what's about to change, so nothing breaks unexpectedly.

For your team

Live codebase graph

Every file as a node, every import as an edge, every function call relationship visible. Open it in any browser. Understand any codebase in five minutes with AI descriptions on every node.

Session history & attribution

Every session logged. Every file touched, attributed to a human or a specific AI agent, grouped by task. AI-powered summaries of what each session accomplished. Full audit trail.

Auto-bootstrap

Mycelium writes CLAUDE.md and .mcp.json into your project automatically. Agents discover it and connect without manual configuration.

The full loop in three calls.

Your team draws the plan on the canvas. Your agent checks what's ready, calls preflight to get the right files, does the work, and marks it done. Your team sees the canvas update and the history fill in. All local. All yours.

01 — agent reads the plan
$curl localhost:47821/roadmap
 "features": [
  { "name": "API Layer",
    "status": "in-progress",
    "dependsOn": ["auth","db"] },
  { "name": "Checkout",
    "status": "planned" },
// agent knows what's active
// and what's blocked
02 — agent gets context
$curl "localhost:47821/preflight?task=build+api+layer"
 "tokensSaved": 7200,
 "files": [
  { "nodeId": "lib/api/router.ts", "score": 0.96 },
  { "nodeId": "lib/auth/middleware.ts", "score": 0.91 },
  { "nodeId": "db/schema.ts", "score": 0.88 },
  { "nodeId": "types/api.ts", "score": 0.84 }
// 4 files. agent starts writing.
03 — team sees everything
$curl localhost:47821/history
 claude-code  lib/api/router.ts    14:32
 claude-code  lib/auth/middleware.ts 14:33
 claude-code  db/schema.ts         14:35
 human       types/api.ts         14:52
// feature marked complete.
// canvas updates. loop closes.

One command. The loop runs forever.

01 — Scan
$ mycelium init

Map everything once

Mycelium parses every file, resolves every import, and builds a full dependency graph. Claude Haiku writes a plain English description of every file. Done in minutes. Stays current automatically.

02 — Plan
$ mycelium start

Open the canvas

Your browser opens the roadmap canvas. Draw your features, set dependencies, find your critical path. Your agents connect to the same MCP server and can read everything you've planned.

03 — Build
GET /preflight?task=

Let the loop run

Your agent reads the plan, calls preflight, works on the right files, marks features complete. You watch the canvas update and the history fill in — with full attribution on every change.

Built for teams that ship with AI and want to stay in control.

Developers

Your agent reads 4 files and gets it right.

Stop watching Claude Code open 30 files before writing a line. Mycelium gives it a map and a plan. Focused agents ship faster and break less.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Any MCP-compatible agent
Engineering leads

You plan it. You see exactly what they built.

Set the roadmap, define priorities, find the critical path. Then watch your agents execute — with full session attribution and a change history that never lies.

  • Shared planning canvas
  • Per-agent attribution
  • AI session summaries
  • Complete audit trail
New engineers

Understand any codebase in five minutes.

Open the graph viewer. Every file, every dependency, every connection — with plain English on every node. The fastest onboarding tool you didn't know you needed.

  • Visual dependency map
  • AI descriptions on every file
  • Semantic search by concept
  • Works on any language

Three ways in. All of them work.

Mycelium works for a solo developer exploring a codebase, an AI agent navigating it autonomously, and a team running both together. Pick the one that fits where you are today.

Human only

You, exploring a codebase you don't know yet.

Run one command. Open a browser. Your entire codebase is mapped as a force-directed graph — files as nodes, imports as edges. Click any file for an AI description of what it does. Search by concept instead of filename. Understand how things connect before you touch anything.

Run mycelium init once to scan and describe every file
Open the graph viewer — click any node to see what imports it and what it imports
Search "authentication flow" and get the right files, not a list of files with "auth" in the name
Plan your work on the roadmap canvas before writing a line
Agent only

Your agent, navigating your code without hand-holding.

Your agent connects to Mycelium's MCP server and calls preflight before touching anything. It gets exactly the files it needs — scored, described, connected. It checks dependencies before modifying a function. It moves fast because it actually knows where it's going.

Calls /preflight?task= before writing a single line — 4 files, not 40
Calls /dependencies?file= to understand what a file imports and what imports it
Calls /xref?file=&fn= to find every caller of a function before modifying it
Every file it touches is logged automatically — the team sees it in the History tab
Human + Agent

Your team plans it. Your agents build it. The loop runs itself.

You draw features on the canvas, set dependencies, find the critical path. Your agent reads that exact canvas through the MCP server and knows what to build next. When it's done, it marks the feature complete. You see the node turn green. The History tab fills in. Nothing is invisible.

You plan: open the canvas, add features, draw deps, hit ● Path to find your critical chain
Agent reads /roadmap — sees what's active, what's blocked, what's ready to build
Agent calls preflight, works on the right files, then PATCHes the feature to complete
You see the canvas update in real time. History shows every file touched, by whom, when

Your team plans it.
Your agents build it.
You see everything.

One command. Any project. The loop between your team and your AI agents — closed.

$ npx @kopikocappu/mycelium init