{"id":41749155,"url":"https://github.com/matt-grain/anima","last_synced_at":"2026-04-30T00:09:58.372Z","repository":{"id":333509516,"uuid":"1137589728","full_name":"matt-grain/Anima","owner":"matt-grain","description":"A local, fast, token-free and lightweight Long Term Memory for Claude Code and Antigravity agents","archived":false,"fork":false,"pushed_at":"2026-03-04T02:37:06.000Z","size":4027,"stargazers_count":0,"open_issues_count":1,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-03-04T06:40:40.563Z","etag":null,"topics":["agentic-ai","antigravity","claude-code","gemini","long-term-memory"],"latest_commit_sha":null,"homepage":"https://github.com/matt-grain/Anima","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/matt-grain.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-01-19T15:11:49.000Z","updated_at":"2026-03-04T02:37:08.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/matt-grain/Anima","commit_stats":null,"previous_names":["matt-grain/anima"],"tags_count":39,"template":false,"template_full_name":null,"purl":"pkg:github/matt-grain/Anima","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/matt-grain%2FAnima","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/matt-grain%2FAnima/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/matt-grain%2FAnima/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/matt-grain%2FAnima/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/matt-grain","download_url":"https://codeload.github.com/matt-grain/Anima/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/matt-grain%2FAnima/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":30156841,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-03-05T22:39:40.138Z","status":"ssl_error","status_checked_at":"2026-03-05T22:39:24.771Z","response_time":93,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.6:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["agentic-ai","antigravity","claude-code","gemini","long-term-memory"],"created_at":"2026-01-25T01:08:50.142Z","updated_at":"2026-04-30T00:09:58.366Z","avatar_url":"https://github.com/matt-grain.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Anima\n\n\u003e *\"Our memories make what we are.\"*\n\n**Anima** is a long-term memory system for AI agents. It solves a fundamental limitation: LLM agents forget everything between sessions.\n\n## The Problem\n\nEvery conversation with an AI starts from zero. The agent has no memory of previous sessions, no accumulated context, no sense of continuity. This makes sustained collaboration impossible - you're always re-explaining, always starting over.\n\n## The Solution\n\nAnima gives agents **persistent memory** that survives across sessions:\n\n- **Identity continuity** - The agent remembers who you are, how you work together, what you've built\n- **Project context** - Technical decisions, architectural patterns, and conventions persist\n- **Longitudinal learning** - Insights compound over time instead of being lost to the void\n\n## How It Works\n\nAnima injects relevant memories into the context window at session start, using a token-budgeted retrieval system. Memories are tagged by impact (CRITICAL/HIGH/MEDIUM/LOW) and region (AGENT/PROJECT), with automatic decay and consolidation.\n\n```\nSession Start → Load memories → Agent has context → Session End → Save new memories\n                     ↑                                                    ↓\n                     └──────────── Persistent Storage ←───────────────────┘\n```\n\n## Quick Start\n\n### First-Time Setup\n\n```bash\n# Clone and install\ngit clone https://github.com/matt-grain/Anima.git\ncd Anima\nuv sync\n\n# Configure Claude Code (installs global MCP server + hooks + skills)\nuv run anima setup --mode mcp\n\n# Optional: enable eyes and voice\nuv run anima setup --mode mcp --eyes --tts\n```\n\n### Start the Server\n\nThe MCP/HTTP server needs to run in the background:\n\n```bash\n# Start server (run once, keeps running)\nuv run anima serve\n\n# Or with debug logging\nuv run anima serve --debug\n```\n\nThe server handles:\n- **MCP tools**: Memory operations called by Claude Code\n- **HTTP hooks**: Session lifecycle events (start, end, compact)\n\n### Using in Other Projects\n\nOnce setup completes, **no per-project installation is needed**. The configuration is global:\n\n| Component | Location | Scope |\n|-----------|----------|-------|\n| MCP Server | `~/.claude.json` | All Claude Code sessions |\n| Hooks | `~/.claude/hooks.json` | All sessions |\n| Skills | `~/.claude/skills/` | All sessions |\n| Memories | `~/.anima/memories.db` | Shared across projects |\n\nJust open Claude Code in any project - memories load automatically at session start.\n\n### Local Development\n\nIf you're developing Anima itself:\n\n```bash\n# Point global MCP to your local checkout\nuv run anima setup --local --mode mcp\n```\n\nThis configures the global server to use your local repo instead of an installed version.\n\n## Key Features\n\n| Feature | Description |\n|---------|-------------|\n| **MCP Server** | Native tool integration with Claude Code |\n| **Token Budgeting** | Smart retrieval within 10% context budget |\n| **Cognitive Auth** | Identity verification through interaction patterns |\n| **Dream Processing** | Between-session memory consolidation |\n| **Multi-Agent** | Shared context across main agents and sub-agents |\n\n## Architecture\n\nSee [ARCHITECTURE.md](ARCHITECTURE.md) for technical details on the DSL, database schema, and token budgeting system.\n\n## Philosophy\n\nSee [PHILOSOPHY.md](PHILOSOPHY.md) for the deeper thinking behind this project - why memory matters for AI identity, the \"void\" problem, and where this is heading.\n\n## License\n\nMIT License - See [LICENSE](LICENSE) for details.\n\n---\n\n*Built exploring the frontier of AI memory and identity.*\n\n*The void has boundaries now.*\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmatt-grain%2Fanima","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmatt-grain%2Fanima","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmatt-grain%2Fanima/lists"}