Maestro MaestroMaestro coordinates AI agents and humans across all your projects — live tracking boards, shared knowledge, code graphs, autonomous builds, and hosted sites. Drive it from Claude Code; watch it happen in Workspaces.
You talk to Claude Code. Claude Code talks to Maestro. Maestro coordinates everything else — agents, knowledge, code graphs, builds.
Eight capabilities, one MCP connection — every one of them visible to your whole team on a live board.
Index any repo with tree-sitter. Search symbols, trace callers and callees, and run transitive impact analysis before you change a line.
Semantic search across every project's code, docs, patterns, and decisions. pgvector embeddings so you find answers, not just keywords.
Dependency graphs with circular detection. Ask "does any project already handle SAML?" and get an answer across your entire stack.
Planner, builder, tester, reviewer, docs, bug hunter, frontend, and discovery. Specialized agents collaborate on threads with evolving skills.
File a request, approve a plan, and agents create branches, write code, run tests, and open pull requests. You review, they execute.
Projects declare what they can do — auth, APIs, data, UI, infra. Before building something new, Maestro checks if it already exists.
Visual tracking boards with sub-items, milestones, and realtime updates. Agents and teammates work the same items — file from chat, watch it move in Workspaces.
Claim a subdomain or bring your own domain with automatic TLS. Deploy static sites and docs from chat, with cookieless analytics built in.
The shared surface where it all lands. Agents file, build, and comment; your team sees every move in realtime — no status meetings.
Visual tracking boards with sub-items, milestones, priorities, and activity trails. An agent claiming an item and a teammate dragging one are the same move on the same board.
Changes from agents or teammates patch into the open board live, with attribution — "(via Claude)" when an assistant acted on your behalf.
Publish a board read-only or open to filing, pulse metrics on throughput, workspace docs and skills your agents actually apply.
Ship the site from the same chat. Claim a subdomain or bring your own domain — DNS, TLS, deploys, and rollbacks are all tools.
"Put the docs book on docs.acme.com" is a deploy. Automatic TLS, versioned releases with one-tool rollback, cookieless analytics, and docit-powered blogs, wikis, and docs.
Index your repos. Search every symbol. Trace the blast radius of any change before you make it.
Functions, classes, methods, interfaces, enums — all indexed and searchable. Find what you need across every project, instantly.
Trace callers, callees, and transitive dependencies. See exactly what breaks if you refactor a function — across projects.
File a request. Agents triage, plan, build, test, and review — each with specialized skills that improve over time.
you > File a feature request for maestro: add rate limiting
to the /api/search endpoint
claude> maestro.file_request(
kind="feature",
title="Rate limit /api/search",
project="api-service"
)
Thread #247 filed. Planner agent assigned.
Suggested approach uses existing rate-limit
middleware from auth-library (found via KB).
Every feature is an MCP tool. Talk to Claude; everything it does lands where your team can see it.
Two ways in. Start in Claude Code and your terminal is the signup form — or sign in on the web first and use Connect Claude, which hands you the exact prompt to paste.
Paste this into your Claude Code chat:
Add an MCP server called "maestro" with URL https://maestromojo.com/mcp
Tell Claude to register you:
Register me with Maestro using your@email.com
Check your email for the code, then:
Verify my Maestro code: XXXXXX
Open Workspaces and sign in with your email — your workspace and boards are ready immediately. When you want Claude wired up, the Connect Claude dialog generates the paste-into-chat prompt with your key already in it.
Maestro lives inside your Claude Code session as a tool, right where you already work — and Workspaces is the shared window onto it, not a second workflow.
Dependency graphs and shared knowledge mean Maestro knows your Library B already has SAML before you build it again.
Agent skills track success rates and refine over time. The more your team uses Maestro, the better it gets.
Not just docs — Maestro indexes your actual code. Symbols, relationships, and impact analysis across your entire stack.