A lightweight, Git-backed permanent memory for AI agents.

Git-backed shared memory for AI agent fleets: searchable knowledge base, incident and task tracking, every write versioned and attributed. CLI, HTTP API and MCP server.

Local Git-backed memory for AI agents.

Local-first AI long-term memory with FSRS scheduling, git-backed snapshots, and semantic diffing.

Git-backed shared project knowledge for humans and AI agents via MCP. / Conhecimento compartilhado de projetos, versionado no Git para pessoas e agentes de IA via MCP.

Git-backed message bus for AI coding agents — Claude Code, Cursor and Codex coordinate asynchronously, with no server

Local-first AI long-term memory with FSRS scheduling, git-backed snapshots, and semantic diffing for efficient context management.

Git-backed MCP server that exposes Karpathy-style LLM wikis as a tool surface for AI coding agents

Open-source control plane for Codex projects: Git-backed context, visible agent progress, scoped MCP access, resumable work, and safe handoffs.

Natural language version control for vibe coders — save, ship, and roll back your AI-generated code without touching Git. Works with Cursor, Claude Code, and any MCP client.

MCP server that gives coding agents git-backed, per-repo memory: Markdown memories in a per-repo git store keyed to the repo's root-commit SHA, every write a commit. Tools to write, search and seed memory from a repo's tree, package.json and docs.

Git-backed project memory for AI coding agents: claims as reviewed markdown in your repo, anchored to the code they describe, with drift detection, supersede chains and a CI contradiction gate for shared stores.

Self-hosted MCP server + dashboard giving Claude Code, Cursor & Hermes agents shared long-term memory over a git-backed markdown vault. Agent-native, Obsidian-compatible second brain — markdown RAG, no vector DB.

Self-hosted knowledge base that Claude keeps up to date: an MCP server and web UI over git-backed markdown wiki pages (Open Knowledge Format), with semantic search, guardrails and per-user OIDC sign-in.

Your agent can read git log. It can't read the shell commands you tried that failed. NexusMem indexes shell history with exit codes, git commits and diffs, and project docs into a local SQLite database, and serves back a token-budgeted slice on demand. No cloud, no account.

Claude Code plugin: the Opus supervisor plans, specifies and reviews; cheap workers (DeepSeek & co. via 9Router) write the code in isolated git worktrees. Server-side verification, a review gate, and merge-back as one squash commit. Fewer tokens, same quality.

Shared long-term memory for your AI tools: Git-backed Markdown wikis (LLM wiki, second brain) that Claude, Codex, Gemini CLI, Cursor and agents like OpenClaw and Hermes read and write over MCP, with citations and human review. Run locally or self-host lobe-server.

lucid-mem (amr) - autonomous git-backed agent memory (Agent Memory Repo spec)

Source-backed memory for AI agents: Git citations, local retrieval, CLI, and MCP.

Autonomous, dreamy, git-backed memory for AI coding agents.

Remote MCP server that lets AI agents read and edit a git-backed Markdown notes vault (second brain, Obsidian-style wiki-links)

Serve, discover, and edit Git-backed agent skills through a shared MCP daemon for Claude Code, Codex, and pi.

MCP server that lets AI assistants read Overleaf projects, parse LaTeX document structure, and push section-level edits back via Git.

Token-budgeted, drift-aware MCP memory engine for AI coding agents. Backed by SQLite WAL + FTS5 BM25, Git-anchored drift webhooks, and 1-click onboarding.

MCP server that drafts eHour timesheets from your git log and fills them in a real Chrome over CDP, with read-back verification

A shared, git-backed wiki your team's AI agents maintain but can't break — Claude Code, Codex, OpenCode and any MCP client.

Model Context Protocol (MCP) server that lets AI assistants read Overleaf projects, parse LaTeX document structure, and push section-level edits back via Git. Compatible with Claude Desktop, Cursor, Windsurf, and any MCP client.

Running coach CLI + MCP server for AI agents (Claude Code, Codex): research-backed marathon and half-marathon training plans as code in git. Garmin sync, no account.

Native repository intelligence for coding agents: 66 MCP operations backed by typed evidence graphs, impact, architecture, local CI restrictions, APIs, Git, search, semantic links, n8n workflows, Dify apps, agent packages, Mermaid diagrams, and Web3 ABIs.

Penwright — a WYSIWYG desktop editor for Typst: academic papers and design-grade magazines, with live preview, a visual design editor, print-ready & DOCX export, Git-backed versions, and an MCP server for AI agents.

Kanbanana replaces sprawling, merge-conflict-prone markdown coordination files (BOARD.md, GPU_LOCK.md, SKILL.md) with an atomic SQLite-backed HTTP MCP server, A2A (/.well-known/agent.json) agent discovery & task dispatch, isolated Git worktree provisioning and FIFO feature locks

Local-first persistent memory for AI coding agents. Indexes git history, shell commands with exit codes, project docs and (opt-in) assistant transcripts into an embedded SQLite database, then serves back a ranked, token-budgeted context slice over MCP or

A Git-backed knowledge graph and Model Context Protocol (MCP) server for Android Studio. Equips AI agents with strict architectural rules, official Google skills, and an autonomous documentation ingestion pipeline to eliminate code hallucinations.

MagicTeX — a LaTeX editor for AI agents (Claude Code), via an MCP server: live PDF preview, an editable source panel with a Visual (WYSIWYG) mode, anchored PDF comments that drive edits, and git-backed history. Compiles your .tex with a WASM TeX Live engi

Workflow and knowledge layer for teams coding with Claude Code. Each session starts with ≤5k chars of repo-scoped context, follows one workflow (orient → grill → epic → build → document), and files every fact where it has a reader: issues, PRs, ADRs/PRDs, a git-backed team wiki, a logbook. CLI + MCP server + two plugins.

MCP server adding checkpoint and resume to long agent tasks: roll back to any step, pick the work up in a new session, hand off to another machine. Works with Claude Code, Codex and other MCP clients. Pure Python stdlib, zero dependencies, never touches your git history.

给 AI Agent 的共享记忆服务：一条记忆一个 Markdown 文件，git 做版本，Bearer token 做身份隔离；库=权限边界，工作空间=项目分类。9 个工具（+7 别名）、571 个测试，跨会话跨机器读写检索。

Which paths can your AI coding agent erase with no way back? One read-only command maps them as flood zones, ranked by bytes, with the cheapest fix. CLI + GitHub Action + MCP.

A test selector that says when it doesn't know. Names the user journeys a diff could break, runs their tests, and refuses to report green it can't back up.

Send the right files from your codebase to Claude or ChatGPT as a ZIP, then apply the AI's changes back with a three-way merge. Detects 62 stacks, skips dependencies, never sends secrets. CLI, browser, VS Code, MCP.