Skip to content
Back to search

mind

Persistent local memory for AI coding agents. Preserve decisions and project context across sessions, compaction, and tools.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked yesterday
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/gabrielmartinmoran--mind/badge.svg)](/listings/gabrielmartinmoran--mind)

HTML

<a href="/listings/gabrielmartinmoran--mind"><img src="/api/listings/gabrielmartinmoran--mind/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

These steps run a script from the network that findsafeskills does not scan.

On macOS or Linux, run:

curl -fsSL https://raw.githubusercontent.com/GabrielMartinMoran/mind/main/scripts/install.sh | bash

The installer installs to user-local paths and installs [Bun](https://bun.sh/) if needed. It doesn't require `sudo`. If your shell can't find `mind`, add `~/.local/bin` to your `PATH` as instructed by the installer, then open a new terminal.

[Install from source and advanced configuration](docs/configuration.md)

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 194 lines · 1 finding marked
  1. ![Mind banner: connected memory nodes](./assets/image/mind-banner.png)
  2. <div align="center">
  3. # 🧠 Mind
  4. **Your coding agent forgets. Mind remembers.**
  5. Local, persistent memory for AI coding agents. Keep decisions, discoveries, fixes,
  6. and working context available across sessions, context compaction, and tools.
  7. **Local-first · Open source · MCP**
  8. [![CI](https://github.com/GabrielMartinMoran/mind/actions/workflows/main.yml/badge.svg?branch=develop)](https://github.com/GabrielMartinMoran/mind/actions/workflows/main.yml)
  9. [![Release](https://img.shields.io/github/v/release/GabrielMartinMoran/mind)](https://github.com/GabrielMartinMoran/mind/releases)
  10. [![License: MIT](https://img.shields.io/github/license/GabrielMartinMoran/mind)](LICENSE)
  11. [![MCP](https://img.shields.io/badge/MCP-supported-6C47FF)](docs/agent-integration-matrix.md)
  12. [Get started](#get-started) · [Documentation](docs/README.md) ·
  13. [Supported agents](docs/agent-integration-matrix.md) · [Benchmarks](#benchmarks)
  14. </div>
  15. ## Contents
  16. - [Why Mind?](#why-mind)
  17. - [Benchmarks](#benchmarks)
  18. - [See Mind in action](#see-mind-in-action)
  19. - [Get started](#get-started)
  20. - [How Mind works](#how-mind-works)
  21. - [Documentation](#documentation)
  22. - [Contributing](#contributing)
  23. ## Why Mind?
  24. Most coding agents lose useful context when sessions end, context is compacted,
  25. or you switch tools. Mind gives them one local memory layer to carry knowledge forward.
  26. - **Remember what matters.** Preserve decisions, constraints, bug fixes, patterns,
  27. and current references instead of rediscovering them every session.
  28. - **Continue unfinished work.** Checkpoints and session summaries help agents
  29. pick up where they left off, including after compaction.
  30. - **Connect knowledge.** Spaces, tags, directional links, and three memory tiers
  31. keep related information organized and useful as a project evolves.
  32. - **Share context across tools.** The CLI, MCP server, HTTP API, and web app
  33. work with the same SQLite database.
  34. - **Search without an API key.** Full-text retrieval works out of the box;
  35. OpenAI or OpenRouter embeddings are optional for semantic search.
  36. ## Benchmarks
  37. ### 🏆 86.67% answer accuracy. Top-3 territory. Even without embeddings.
  38. Mind achieved **86.67% answer accuracy** on
  39. [OmniMemEval LongMemEval-S](https://github.com/MemTensor/OmniMemEval),
  40. putting it in **top-three territory by raw score** against
  41. [published results](https://github.com/MemTensor/OmniMemEval/blob/main/docs/user_memory/results.md)
  42. — with semantic RAG disabled.
  43. **But remembering the right answer is only one part of memory.**
  44. Mind is designed to build, connect, prioritize, and maintain knowledge over time:
  45. promote useful memories, preserve superseded decisions, link related discoveries,
  46. and recover the context needed for the next session.
  47. _Benchmark note: Mind's completed run used a stratified subset and GPT-6 Luna
  48. (low) for answering and judging, while the published comparisons use different
  49. sample sizes and configurations. The top-three comparison is indicative, not an
  50. official ranking. RAG was disabled; additional RAG-enabled tests produced similar
  51. results but are not a controlled head-to-head comparison._
  52. [Mind's benchmark methodology and results](benchmarks/omnimemeval/README.md) ·
  53. [How the memory lifecycle works](docs/cognitive-memory-lifecycle.md)
  54. ## See Mind in action
  55. ![Mind web app preview](./assets/video/web-preview.gif)
  56. Explore your memories in the web app, edit what matters, and follow
  57. relationships in the per-space Neural Map.
  58. ## Get started
  59. ### 1. Install Mind
  60. On macOS or Linux, run:
  61. ```bash
  62. curl -fsSL https://raw.githubusercontent.com/GabrielMartinMoran/mind/main/scripts/install.sh | bash
    • Note: Pipes a downloaded script directly into a shell (curl | sh / wget | bash)
    • Line 89 on GitHub
  63. ```
  64. The installer installs to user-local paths and installs [Bun](https://bun.sh/)
  65. if needed. It doesn't require `sudo`. If your shell can't find `mind`, add
  66. `~/.local/bin` to your `PATH` as instructed by the installer, then open a
  67. new terminal.
  68. [Install from source and advanced configuration](docs/configuration.md)
  69. ### 2. Connect your agents
  70. ```bash
  71. mind setup
  72. ```
  73. This configures **all supported agent integrations** and the shared
  74. `mind-management` skill. It does not install the agent applications themselves
  75. and may write configuration for agents that are not installed.
  76. Only want one integration? Use `mind setup <agent>` (for example,
  77. `mind setup codex`). Run `mind setup compatibility` to see supported,
  78. partial, and unverified capabilities, or `mind setup refresh` to update
  79. existing Mind-managed integrations.
  80. [Supported agents and capability matrix](docs/agent-integration-matrix.md) ·
  81. [Agent setup details](docs/agent-setup.md)
  82. ### 3. Try it
  83. ```bash
  84. mind status
  85. mind create projects/my-project "My project"
  86. mind add projects/my-project architecture "CLI uses command registry + atomic command modules"
  87. mind search architecture
  88. mind serve start
  89. ```
  90. Open [http://localhost:30303](http://localhost:30303) to inspect your memories.
  91. Agents use Mind through its installed MCP integration; the commands above are
  92. a quick way to explore the same store yourself.
  93. **Using an agent to set up Mind?** Give it the installation steps above and ask
  94. it to run `mind setup`, check `mind status`, and report
  95. `mind setup compatibility` results. Your agent should distinguish configured
  96. integrations from those actually validated on your machine.
  97. ## How Mind works
  98. Mind is a persistent cognitive workspace, not just a collection of notes.
  99. 1. **Remember.** Save reusable decisions, fixes, discoveries, and references;
  100. use checkpoints for unfinished work and summaries for completed sessions.
  101. 2. **Recall.** Retrieve relevant context through Mind's unified full-text search,
  102. with optional embeddings when semantic search is enabled.
  103. 3. **Relate and evolve.** Link memories, maintain current references, preserve
  104. superseded knowledge, and let useful memories move between active and archive
  105. tiers as their use changes.
  106. 4. **Continue.** Agents follow the shared `mind-management` skill to recover
  107. context and manage memory across sessions and handoffs, subject to their
  108. integration's capabilities.
  109. Everything lives in one local SQLite store (`mind.db`). Mind also supports
  110. an **experimental** opt-in [Markdown autosync](docs/autosync.md) for versioned
  111. project memories. Optional embedding providers make external API calls only
  112. when semantic search is enabled.
  113. [Memory lifecycle](docs/cognitive-memory-lifecycle.md) ·
  114. [Project philosophy](docs/mind-philosophy.md)
  115. ## Documentation
  116. Looking for something specific? Start with the
  117. [documentation index](docs/README.md):
  118. - [Agent compatibility and setup](docs/agent-integration-matrix.md)
  119. - [CLI, web UI, and MCP reference](docs/interfaces.md)
  120. - [Search engine and query syntax](docs/search-engine.md)
  121. - [Autosync (experimental)](docs/autosync.md)
  122. - [Configuration and optional semantic search](docs/configuration.md)
  123. - [Skill-first agent protocol](docs/skill-first-distribution.md)
  124. For all available CLI commands, run `mind help`.
  125. ## Contributing
  126. Issues and pull requests are welcome. Read [CONTRIBUTING.md](CONTRIBUTING.md)
  127. for contribution guidelines, tests, and quality gates, and
  128. [Mind's philosophy](docs/mind-philosophy.md) for the project's principles.
  129. ## License
  130. [MIT](LICENSE) · [Changelog](CHANGELOG.md)
  131. <details>
  132. <summary>Star history</summary>
  133. <a href="https://www.star-history.com/?repos=gabrielmartinmoran%2Fmind&type=date&releases=&legend=bottom-right">
  134. <picture>
  135. <source media="(prefers-color-scheme: dark)" srcset="https://api.star-history.com/chart?repos=gabrielmartinmoran/mind&type=date&theme=dark&legend=bottom-right" />
  136. <source media="(prefers-color-scheme: light)" srcset="https://api.star-history.com/chart?repos=gabrielmartinmoran/mind&type=date&legend=bottom-right" />
  137. <img alt="Star History Chart" src="https://api.star-history.com/chart?repos=gabrielmartinmoran/mind&type=date&legend=bottom-right" />
  138. </picture>
  139. </a>
  140. </details>

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.