PMB
For AI coding users: PMB gives Claude, Cursor, and Zed persistent project memory so agents understand your codebase without re-explanation.
Built by Oleksii Bondar · Artificial Intelligence · Developer Tools · Open Source · Free
In category: #311 of 2048 · Artificial Intelligence
PMB is a Product Hunt listing on Launch Llama with 225 total upvotes, founded by Oleksii Bondar who joined in Jul 2026, compared against 6 alternatives.
Product Hunt listing
Listing status
225upvotes
Community score
Oleksii Bondar (Joined Jul 2026)
Founder
Compared against 6 alternatives
Market comparison
Launched on Launch Llama Jul 6, 2026 · #22 of 53 that day
About
PMB gives Claude Code, Cursor, Codex and Zed persistent project memory through MCP. It stores decisions, lessons, goals, recent work, project facts and docs in one SQLite workspace on your disk. No cloud, no API keys, no LLM call on the read path. It is open source, offline-first, inspectable/exportable, with a local dashboard and honest impact tracking so you can see which memories actually help.
Competitors: Codebase Context - MCP-based memory, Continue.dev - IDE memory & context, Aider - AI pair programming memory, CodeCompanion - persistent code context, Repo Prompt - codebase summarization
Launch Llama's Take
Based on Launch Llama data as of
PMB has 225 upvotes, including the votes it received on Product Hunt, which ranks it #311 of 2,048 Artificial Intelligence tools on Launch Llama by upvotes, in the top quarter. It has been listed on Launch Llama since July 2026. No Launch Llama members have upvoted it yet and it has no reviews yet.
PMB vs ContextPool
PMB stores decisions, lessons, and project facts in a local SQLite workspace, while ContextPool extracts insights automatically from past coding sessions.
| PMB | ContextPool | Universal Memory Tool | Agentmemory | |
|---|---|---|---|---|
| What it does | For AI coding users: PMB gives Claude, Cursor, and Zed persistent project memory so agents understand your codebase without re-explanation. | For AI coding teams: ContextPool gives your agent persistent memory by extracting bugs, fixes, and design decisions from past sessions an… | For AI developers: Universal Memory Tool compresses chats and documents into searchable memory for your AI assistant so it recalls contex… | For developers: Agentmemory gives Claude Code and coding agents searchable persistent memory, reducing token usage by 92% while keeping 1… |
| Pricing | Free | Freemium | Paid | Free |
| Free plan | Yes, free to use | Yes, free plan | No free plan listed | Yes, free to use |
| Category | Open Source | Open Source | Artificial Intelligence | Open Source |
| Upvotes on Launch Llama | 225 | 174 | 6 | 253 |
| Listed on Launch Llama | July 2026 | April 2026 | May 2026 | May 2026 |
Key Features
- Store decisions, lessons, and project facts in a local SQLite file
- Automatically recall relevant memory before the model responds
- Track which lessons the agent actually follows and flags unused ones
- Visualize memory as an interactive graph and timeline
- Work offline with no API keys or cloud dependencies
Use Cases
- Set up PMB's SQLite workspace, document architectural decisions and lessons learned for Claude/Cursor agent reference
- Use PMB's local dashboard to track which stored memories (decisions, goals, recent work) actually reduce agent errors and improve code generation quality
- Export and inspect your project memory file to audit what context the AI agents have access to and remove sensitive information before sharing codebases
- Store project facts, documentation links, and refactoring decisions in PMB for Zed's AI to maintain consistency across files
- Connect PMB via MCP to Claude/Cursor workflow and monitor whether persistent memory reduces clarification questions
Who it is for
- Developers who use AI coding agents like Claude Code, Cursor, or Codex on ongoing projects
- Engineers who want local-first, offline memory for their AI coding tools
- Teams sharing one workspace memory across several AI coding agents
Pricing
PMB is free to use. See the website for details.
FAQ
Does my code or data ever leave my machine?
No. Everything lives in a local SQLite file with vectors in LanceDB right next to it. There are no network calls on the read path, no account and no telemetry, ever. Unplug the internet and it still works.
How is this different from RAG or a vector database?
Two ways. Recall is hybrid, BM25 plus dense vectors plus an entity graph, fused and ranked. And it's automatic: the right memory is injected before the model thinks. You don't build a pipeline or hope the agent remembers to call a tool.
Will it slow my agent down?
No. Recall lands in about 35 ms and writes return in under a millisecond, the embedding and vector insert happen on a background thread, so the turn is never blocked.
Which agents and operating systems are supported?
Any MCP-aware agent: Claude Code, Cursor, Codex, Zed, Windsurf and more, wired in with one command. PMB is pure Python and tested on macOS, Linux and Windows.
What if a memory is wrong or unhelpful?
PMB scores whether each lesson actually gets followed and flags the dead ones so you can prune them. It's the rare tool that tells you when its own memory isn't earning its place.
Is it really free?
Yes. Apache-2.0, open source, free forever. No paid tier, no seats, no telemetry. You own the file and the code.
Alternatives
More in Artificial Intelligence
New on Launch Llama
Launched this week · This month's launches · All products A–Z
Ask AI about PMB
Third-party sources
- Show HN: PMB – local memory for coding agents that shows if it is used
Hacker News · June 29, 2026 · 26 points
For bots & AI agents
Machine-readable versions of PMB and Launch Llama. Fetch these instead of parsing this page.
This listing (.md) llms.txt llms-full.txt ai.txt Full catalog (.md) Endpoint index API docs MCP server MCP manifest Developer docs