basic-memory
Basic Memory is an open-source AI memory system that enables LLMs (Claude, ChatGPT, Cursor) to maintain persistent knowledge across conversations using local Markdown files. It offers both a free self-hosted option and a paid cloud version with cross-device sync and collaborative features.
Key facts
Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.
| Field | Value |
|---|---|
| Repository | basicmachines-co/basic-memory |
| Owner | basicmachines-co |
| Primary language | Python |
| License | AGPL-3.0 — OSI-approved |
| Stars | 3.4k |
| Forks | 225 |
| Open issues | 104 |
| Latest release | v0.22.1 (2026-06-13) |
| Last updated | 2026-07-08 |
| Source | https://github.com/basicmachines-co/basic-memory |
What basic-memory is
Python-based MCP (Model Context Protocol) server that stores conversations and knowledge as Markdown files, provides semantic search, wikilink graphs, and bidirectional sync. Supports multiple AI clients via stdio/HTTPS transports; cloud version uses Postgres and S3 for hosting.
Get the basic-memory source
Clone the repository and explore it locally.
git clone https://github.com/basicmachines-co/basic-memory.gitcd basic-memory# follow the project's README for install & configurationNeed it deployed, integrated, or customized instead? DEV.co ships production installs.
Best use cases
Implementation considerations
- Local install requires Python 3.12+ and uv tool manager; cloud option eliminates this dependency.
- MCP server architecture requires client-side support (Claude Desktop, Cursor, VS Code, etc.); verify your AI tool supports MCP protocol.
- Markdown-based storage means knowledge graph scaling and query performance may degrade with very large note counts; no published benchmarks available.
- Bidirectional sync between cloud and local disk uses rclone with conflict resolution; test thoroughly before production sync in team settings.
- AGPL copyleft requires audit of derivative works and SaaS deployments; cloud version compliance responsibility unclear from README.
When to avoid it — and what to weigh
- You require non-AGPL licensing for commercial products — AGPL-3.0 requires derivative works and SaaS offerings to be open-source. Verify with your legal team before embedding in proprietary products.
- You need long-term vendor stability guarantees — Project is ~1.5 years old with active development but unknown commercial backing. Evaluate risk tolerance for production dependencies on early-stage OSS.
- Your workflow doesn't use Markdown or MCP clients — Requires integration with Claude Desktop, Claude Code, Cursor, or similar MCP-capable tools. No native support for other AI platforms or non-MCP workflows.
- You need enterprise support SLAs — Cloud offering is beta-stage pricing; no published SLAs, incident response times, or enterprise support tiers mentioned.
License & commercial use
AGPL-3.0 (GNU Affero General Public License v3.0). Source code is open and freely usable for internal non-commercial use. Any modification or network service offering (including SaaS) must make source code available to all users.
AGPL-3.0 is a copyleft license. Using or modifying the code for any commercial product, SaaS offering, or service open to external users legally requires making derivative works open-source and providing source access to all users. Embedding in proprietary enterprise software is not permitted without explicit relicensing. The cloud offering ($15/mo) is a separate commercial service, not an exception to AGPL terms. Requires legal review before any commercial deployment.
DEV.co evaluation signals
Editorial assessment — not user reviews. Directional, with an explicit confidence level.
| Signal | Assessment |
|---|---|
| Maintenance | Active |
| Documentation | Adequate |
| License clarity | Clear |
| Deployment complexity | Low |
| DEV.co fit | Good |
| Assessment confidence | High |
Cloud version uses WorkOS AuthKit for auth, Neon Postgres, and Tigris S3 (unknown security certifications/SOC 2 status). Local install stores all data as plaintext Markdown on disk—requires OS-level file encryption if sensitive. MCP stdio transport runs within local process; HTTPS transport for cloud and remote clients. No audit, penetration test results, or vulnerability disclosure policy mentioned. Requires review of cloud infrastructure security posture before sensitive data storage.
Alternatives to consider
Obsidian with community plugins (Canvas, Graph, AI)
Markdown-first knowledge management with local storage and AI integrations; less MCP-native and requires manual plugin curation.
Mem.ai or Notion AI
LogSeq or Roam Research
Graph-based note-taking with some AI features; not MCP-native, less tightly integrated with Claude/ChatGPT, no official cloud sync.
Build on basic-memory with DEV.co software developers
Try the cloud version free for 7 days (no credit card), or self-host locally for free. Verify MCP client support and AGPL licensing requirements before production use.
Talk to DEV.coRelated open-source tools
Surfaced by semantic similarity across the DEV.co open-source index.
Related on DEV.co
Explore the category and the services that help you build with it.
basic-memory FAQ
Can I use this in a commercial SaaS product?
Does the local install sync across devices?
What happens to my data if I stop paying for Cloud?
Does it work offline?
Software development & web development with DEV.co
Adopting basic-memory is usually one piece of a larger software development effort. As a software development agency, DEV.co provides software development services and web development expertise — pairing senior software developers and web developers with your team to design, build, and operate mcp servers software in production.
Evaluate Basic Memory for Your AI Workflow
Try the cloud version free for 7 days (no credit card), or self-host locally for free. Verify MCP client support and AGPL licensing requirements before production use.