Your AI forgets everything when you close the tab. Phloem fixes that. Phloem is a local MCP server that gives your AI persistent memory across sessions. It works with any tool that supports the Model Context Protocol — an open standard. Today that includes Claude Code, VS Code, Cursor, Windsurf, Zed, Neovim, Cline, Warp, Continue, JetBrains, and more arriving every week. You install it once. Every
Add this skill
npx mdskills install CanopyHQ/phloemMCP server providing persistent memory with semantic search, code citations, and causal graphs
Your AI forgets everything when you close the tab. Phloem fixes that.
Phloem is a local MCP server that gives your AI persistent memory across sessions. It works with any tool that supports the Model Context Protocol — an open standard. Today that includes Claude Code, VS Code, Cursor, Windsurf, Zed, Neovim, Cline, Warp, Continue, JetBrains, and more arriving every week.
You install it once. Every MCP-compatible AI tool on your machine gets memory.
brew install canopyhq/tap/phloem # or download binary from Releases
phloem setup # auto-detects your tools
.claude Files or Markdown?Every AI coding tool has some form of project context — .claude/ directories, CLAUDE.md, .cursorrules, system prompts. They work. So why do you need Phloem?
They're siloed. Your .claude files don't help Cursor. Your .cursorrules don't help Claude Code. Switch tools and you start from zero. Phloem is one memory that works across all of them because MCP is a shared standard.
They're static text. You write rules by hand and hope the AI reads them. Phloem memories are created naturally during conversation, found via semantic search (not keyword matching), and linked to the code they reference.
They can't track what changed. You wrote "we use JWT with 15-minute expiry" in a markdown file. Then someone changed it to 30 minutes. The file still says 15. Phloem citations link memories to specific lines of code and automatically decay confidence when the code drifts. Your AI never confidently cites stale information.
They don't capture why. Markdown files tell the AI what to do. Phloem's causal graph captures the chain of decisions — why you chose JWT over sessions, what that caused you to change, and what would break if you reverted it.
You: "Add the logout endpoint"
Phloem finds "JWT tokens expire in 15 minutes, refresh tokens in 7 days" — even though nothing in your query mentions JWT or tokens. It understands meaning, not just words.
Memory: "Rate limiter uses sliding window, 100 req/min per API key"
→ src/middleware/rate_limit.go:42-67 (confidence: 0.95)
Refactor that file? Confidence decays. Delete it? Citation marked invalid. Your AI adapts.
"Switched from REST to gRPC"
→ caused: "Updated service clients to protobuf"
→ caused: "Added buf.gen.yaml to build pipeline"
→ caused: "Rewrote integration tests for streaming"
Ask "what would break if we reverted gRPC?" and get a real answer.
Zero network requests. Not as a policy — there is no networking code in the binary. Your memories are a SQLite file on your disk. phloem audit proves it.
Phloem speaks MCP — the same open protocol that Claude Code, VS Code, Cursor, Zed, and others already support. Add a new AI tool? If it speaks MCP, it already has access to your full memory.
brew install canopyhq/tap/phloem
phloem setup claude-code
No restart needed. Start a session and your AI has memory.
For all other supported tools (VS Code, Cursor, Windsurf, Zed, Neovim, Cline, Warp, Continue, JetBrains), see the Supported Tools Guide.
Or just run phloem setup — it auto-detects everything on your machine.
macOS (Homebrew):
brew install canopyhq/tap/phloem
phloem setup
Windows / Linux:
Download the binary for your platform from GitHub Releases, extract it, add to PATH, then run phloem setup.
From source:
git clone https://github.com/CanopyHQ/phloem.git && cd phloem
CGO_ENABLED=1 go build -o phloem .
./phloem setup
Requires Go 1.24+ and a C compiler (for sqlite-vec/CGO).
SQLite + sqlite-vec — Everything in ~/.phloem/memories.db. Vector embeddings power semantic search. No external services.
Causal DAG — Memories linked by cause and effect. Your AI traverses the graph to understand full chains of reasoning.
Citation verification — Memories attach to file:line ranges. When code drifts, confidence decays automatically.
MCP Protocol — JSON-RPC over stdio. No HTTP server, no ports, no network surface. Any MCP client connects instantly.
There is no networking code in the binary. Verify it:
phloem audit
sudo lsof -i -P | grep phloem # should show nothing
No accounts. No telemetry. No analytics. Delete anytime: rm -rf ~/.phloem. Full details →
We welcome contributions. See CONTRIBUTING.md for guidelines.
Apache 2.0 License. Built by Canopy HQ LLC.
Install via CLI
npx mdskills install CanopyHQ/phloemPhloem is a free, open-source AI agent skill. Your AI forgets everything when you close the tab. Phloem fixes that. Phloem is a local MCP server that gives your AI persistent memory across sessions. It works with any tool that supports the Model Context Protocol — an open standard. Today that includes Claude Code, VS Code, Cursor, Windsurf, Zed, Neovim, Cline, Warp, Continue, JetBrains, and more arriving every week. You install it once. Every
Install Phloem with a single command:
npx mdskills install CanopyHQ/phloemThis downloads the skill files into your project and your AI agent picks them up automatically.
Phloem works with Claude Code, Claude Desktop, Cursor, Vscode Copilot, Windsurf, Continue Dev, Codex, Gemini Cli, Amp, Roo Code, Goose, Opencode, Trae, Qodo, Command Code. Skills use the open SKILL.md format which is compatible with any AI coding agent that reads markdown instructions.