the ecosystem
a suite of open-source tools for ai-powered development.
mcp servers, libraries, and infrastructure — all built in Rust.
octomind
ai agent runtime — one binary, any model, any domain
open-source runtime for specialist ai agents. a single Rust binary that runs specialist agents with 20+ ai providers, mcp-native tooling, adaptive compression, and daemon mode. install agents, not frameworks.
tap
community registry of specialist ai agents
the open registry of production-ready ai agents, capabilities, and skills. browse 150+ specialist agents across development, law, medicine, finance, and security. contribute your domain expertise as installable TOML manifests.
octocode
structural code intelligence — search, graph, lsp
two-tier code intelligence for ai agents. a live ast graph — files, symbols, imports, calls, extends, implements — built by tree-sitter with zero indexing, always on. layer in optional semantic search and lsp-precise go-to-definition when you need them. retrieval quality is a published, reproducible benchmark, not a claim: hit@10 0.992 on code search.
octobrain
persistent memory and knowledge base for ai, with proof
give your ai assistant a long-term memory that outlives the context window — 22 memory types, hybrid BM25 + vector search, and a sleep-consolidation pass that folds a finished goal into one summary instead of leaving a hundred loose notes behind. retrieval quality is a measured, reproducible BEIR benchmark, not a claim.
octofs
filesystem mcp server with ssh-native reach
the filesystem engine every octomind agent runs on — and a standalone mcp server anyone can point at Cursor, Claude, or Windsurf. hash-verified line edits so a stale view can never silently corrupt a file, transparent ssh/SFTP on every tool, and a shell that knows the difference between a real command and a call that should have used the dedicated tool. native Rust, zero runtime.
octolib
unified ai provider library in Rust — the engine under octomind
the provider layer octomind and OctoHub are both built on. one type-safe Rust api for 30 ai providers, OAuth into an existing ChatGPT or Claude subscription instead of a new api key, thinking/reasoning tracked separately from content, and a comment-preserving config-migration engine borrowed straight from octomind's own upgrade path.
octohub
high-performance llm proxy — powers hub and cloud
the proxy that sits in front of every model octomind hub and octomind cloud serve — one OpenAI-style endpoint, 20+ upstream providers, and every request recorded in your own database. completion chaining replays prior turns automatically, including reasoning blocks. deploy in minutes: one binary, one config file, schema created on first connection.
octoweb
keyboard-first macOS browser with ai, and a memory
a macOS browser built around the keyboard, not the mouse. the ai sidebar runs a real octomind agent over ACP — not a bolted-on extension — and an opt-in background agent can memorize your browsing every 30 minutes so the sidebar already has context you never typed. mail, notes, a terminal, and Chrome extensions sit beside the page, each scoped to a workspace. the browser is also a 38-tool mcp server: any ai client can navigate, click, fill forms, read the page, and search your mail directly. WebKit + Rust. no Electron. no Chrome.