{"id":51492137,"url":"https://github.com/evermind-ai/raven","last_synced_at":"2026-07-07T12:02:11.395Z","repository":{"id":368545947,"uuid":"1245465729","full_name":"EverMind-AI/Raven","owner":"EverMind-AI","description":"A memory-first agent that gets smarter with every interaction.","archived":false,"fork":false,"pushed_at":"2026-07-01T05:16:10.000Z","size":12504,"stargazers_count":20,"open_issues_count":3,"forks_count":1,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-07-01T05:19:01.305Z","etag":null,"topics":["ai","ai-agents","anthropic","chatgpt","claude","codex","evermind","hermes","hermes-agent","llm","openai","openclaw","openhuman","self-evolving","self-improving"],"latest_commit_sha":null,"homepage":"https://raven.evermind.ai","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/EverMind-AI.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":"CODE_OF_CONDUCT.md","threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":"SECURITY.md","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":"2026-05-21T08:45:07.000Z","updated_at":"2026-07-01T04:53:50.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/EverMind-AI/Raven","commit_stats":null,"previous_names":["evermind-ai/raven"],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/EverMind-AI/Raven","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EverMind-AI%2FRaven","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EverMind-AI%2FRaven/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EverMind-AI%2FRaven/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EverMind-AI%2FRaven/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/EverMind-AI","download_url":"https://codeload.github.com/EverMind-AI/Raven/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EverMind-AI%2FRaven/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":35226918,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-07-07T02:00:07.222Z","response_time":90,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"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":["ai","ai-agents","anthropic","chatgpt","claude","codex","evermind","hermes","hermes-agent","llm","openai","openclaw","openhuman","self-evolving","self-improving"],"created_at":"2026-07-07T12:02:08.658Z","updated_at":"2026-07-07T12:02:11.335Z","avatar_url":"https://github.com/EverMind-AI.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cdiv align=\"center\" id=\"readme-top\"\u003e\n\n![Raven banner](https://github.com/user-attachments/assets/5a99d736-49ee-49c9-8b51-890f14078e78)\n\n\u003cp align=\"center\"\u003e\n  \u003ca href=\"https://x.com/evermind\"\u003e\u003cimg src=\"https://img.shields.io/badge/EverMind-000000?labelColor=gray\u0026style=for-the-badge\u0026logo=x\u0026logoColor=white\" alt=\"X\"\u003e\u003c/a\u003e\n  \u003ca href=\"https://huggingface.co/EverMind-AI\"\u003e\u003cimg src=\"https://img.shields.io/badge/🤗_HuggingFace-EverMind-F5C842?labelColor=gray\u0026style=for-the-badge\" alt=\"HuggingFace\"\u003e\u003c/a\u003e\n  \u003ca href=\"https://discord.gg/gYep5nQRZJ\"\u003e\u003cimg src=\"https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fdiscord.com%2Fapi%2Fv10%2Finvites%2FgYep5nQRZJ%3Fwith_counts%3Dtrue\u0026query=%24.approximate_presence_count\u0026suffix=%20online\u0026label=Discord\u0026color=404EED\u0026labelColor=gray\u0026style=for-the-badge\u0026logo=discord\u0026logoColor=white\" alt=\"Discord\"\u003e\u003c/a\u003e\n  \u003ca href=\"https://github.com/EverMind-AI/EverOS/discussions/67\"\u003e\u003cimg src=\"https://img.shields.io/badge/WeCom-EverMind_社区-07C160?labelColor=gray\u0026style=for-the-badge\u0026logo=wechat\u0026logoColor=white\" alt=\"WeChat\"\u003e\u003c/a\u003e\n\u003c/p\u003e\n\n[Website](https://raven.evermind.ai) · [中文](README.zh-CN.md)\n\n\u003c/div\u003e\n\n\u003cbr\u003e\n\n# Raven\n\nRaven is **The Self-Improving Agent Harness**, built on [EverOS](https://github.com/EverMind-AI/EverOS).\n\nRaven helps agents improve across runs by continuously refining the systems around them: tools, skills, memory, code execution, policies, and working environment. EverOS provides durable user memory, agent memory, and world knowledge across sessions, so successful workflows can evolve into reusable Agent Templates and digital workers.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"https://github.com/user-attachments/assets/a4dc5b21-c8e7-4397-95e1-50afeeb826e4\" alt=\"Starting Raven from the command line\" width=\"100%\"\u003e\n\u003c/p\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003ckbd\u003eTable of Contents\u003c/kbd\u003e\u003c/summary\u003e\n\n\u003cbr\u003e\n\n- [Quick Install](#quick-install)\n- [What You Can Do in 2 Minutes](#what-you-can-do-in-2-minutes)\n- [Messaging Gateways](#messaging-gateways)\n- [Why Raven](#why-raven)\n- [What Raven Is Built For](#what-raven-is-built-for)\n- [Agent Templates](#agent-templates)\n- [Useful Commands](#useful-commands)\n- [Docs by Goal](#docs-by-goal)\n- [Architecture](#architecture)\n- [Developer Workflow](#developer-workflow)\n- [Status](#status)\n- [Star Us](#star-us)\n- [EverMind Ecosystem](#evermind-ecosystem)\n- [Contributing](#contributing)\n\n\u003cbr\u003e\n\n\u003c/details\u003e\n\n## Quick Install\n\n### Linux, macOS, WSL2\n\n```bash\ncurl -fsSL https://raven.evermind.ai/install.sh | bash\n```\n\n### Windows (native, PowerShell)\n\n\u003e **Heads up:** Native Windows runs Raven without WSL. CLI, TUI, gateway, and\n\u003e tools install natively. If you would rather use WSL2, the Linux/macOS\n\u003e one-liner above works there too.\n\nRun this in PowerShell:\n\n```powershell\nirm https://raven.evermind.ai/install.ps1 | iex\n```\n\n### After installation\n\nThe installer handles everything: uv, Python 3.12, Node.js 22, and Raven.\n\nOpen a new terminal. On Linux, macOS, or WSL2, you can also reload your current\nshell:\n\n```bash\nsource ~/.bashrc    # or: source ~/.zshrc\n```\n\nThen run:\n\n```bash\nraven onboard\nraven\n```\n\nRaven supports OpenRouter, OpenAI, Anthropic, Gemini, DeepSeek, GitHub Copilot,\nOpenAI Codex OAuth, and custom OpenAI-compatible endpoints.\n\nIf setup fails or a provider is not ready, run:\n\n```bash\nraven doctor\n```\n\n## What You Can Do in 2 Minutes\n\n- Start the Raven harness in a terminal-native TUI with `raven` or `raven tui`.\n- Run a one-shot shell task with `raven agent -m \"...\"`.\n- Configure providers, sandboxing, channels, and memory with `raven onboard`.\n- Browse built-in and local SkillForge skills with `raven skill list`.\n- Resume, fork, export, or delete previous work with `raven sessions list`.\n- Check proactive memory and scheduled nudges with `raven sentinel status`.\n\n## Messaging Gateways\n\nRaven currently ships 12 gateway adapters. Use `raven channels list` to see the\nadapters available in your local install and `raven gateway` to run the gateway\ndaemon.\n\n| Gateway | Adapter id | Notes |\n| --- | --- | --- |\n| Telegram | `telegram` | Bot-based messaging |\n| Slack | `slack` | Workspace messaging |\n| Discord | `discord` | Server and bot messaging |\n| WhatsApp | `whatsapp` | Uses the bundled TypeScript bridge |\n| Matrix | `matrix` | Matrix rooms and direct messages |\n| Feishu | `feishu` | Lark/Feishu app integration |\n| WeCom | `wecom` | WeCom group and app messaging |\n| Mochat | `mochat` | API/socket-based messaging |\n| QQ | `qq` | QQ bot integration |\n| DingTalk | `dingtalk` | DingTalk stream integration |\n| Email | `email` | IMAP/SMTP mailbox integration |\n| WeChat | `weixin` | Personal WeChat adapter; `weixin` is the current CLI id |\n\n## Why Raven\n\nMost agent tools stop at \"LLM + tools + loop.\" That works for demos, but it\nbreaks down when the agent becomes part of your daily environment:\n\n- Long sessions overflow context and lose important details.\n- Every turn re-sends the same system prompt, skills, and tool definitions.\n- The agent waits passively even when it can see something that needs action.\n- Useful workflows stay trapped in chat history instead of becoming reusable\n  skills.\n\nRaven treats the harness around the agent as the product, not a thin wrapper or\nan edge case.\n\nRaven's self-improving harness is built around three product bets:\n\n- **Memory-first harness:** user memory, agent memory, and world knowledge stay\n  separate, durable, and reusable across sessions.\n- **Self-improving skills:** repeated workflows can become skills, collect\n  feedback, and evolve instead of staying buried in chat history.\n- **Agent Templates:** builders can start from Raven, define an agent for a\n  scenario, and share it without rebuilding the harness layer.\n\n\u003ctable\u003e\n\u003ctr\u003e\n\u003cth width=\"28%\"\u003eCapability\u003c/th\u003e\n\u003cth width=\"36%\"\u003eRaven\u003c/th\u003e\n\u003cth width=\"36%\"\u003eTypical tool-based agent\u003c/th\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eNative terminal product\u003c/strong\u003e\u003c/td\u003e\n\u003ctd\u003eInteractive TUI, CLI, gateway mode, and typed RPC between Python and React/Ink\u003c/td\u003e\n\u003ctd\u003eUsually a thin command wrapper around a chat loop\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eLong memory\u003c/strong\u003e\u003c/td\u003e\n\u003ctd\u003eEverOS-backed memory, local skills, session history, and workspace templates\u003c/td\u003e\n\u003ctd\u003eUsually transient context or provider-side chat history\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eContext control\u003c/strong\u003e\u003c/td\u003e\n\u003ctd\u003eCurator and legacy context engines with explicit token budgets and fail-safes\u003c/td\u003e\n\u003ctd\u003eUsually truncation, summarization, or hidden prompt heuristics\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eProactivity\u003c/strong\u003e\u003c/td\u003e\n\u003ctd\u003eSentinel, scheduler, nudge policy, and deferred decision flow\u003c/td\u003e\n\u003ctd\u003eUsually waits until the user types again\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eSkill evolution\u003c/strong\u003e\u003c/td\u003e\n\u003ctd\u003eDetects reusable procedures, materializes skills, tracks feedback, and evolves them\u003c/td\u003e\n\u003ctd\u003eUsually static markdown prompts or manually installed plugins\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/table\u003e\n\n\u003cbr\u003e\n\n## What Raven Is Built For\n\nRaven is designed for the workflows where ordinary chat agents and static tool\nloops feel too small.\n\n### 1. Terminal-Native Daily Work\n\nRaven can run the harness as a native TUI, a direct CLI entry point, or a\ngateway-backed runtime. The TUI is not a web shell: it is a React/Ink\napplication talking to Raven's Python runtime through a typed RPC protocol.\n\n### 2. Memory That Becomes Useful\n\nRaven connects the harness to EverOS for long-term user and agent memory.\nSessions, procedures, and reusable patterns can be turned into local skill\nmaterial instead of disappearing into old transcripts.\n\n### 3. Context That Does Not Collapse Under Pressure\n\nThe context stack has a legacy path and a Curator path. Under pressure, the\nharness can archive, retrieve, and assemble context with explicit budgets\ninstead of blindly clipping the oldest messages.\n\n### 4. Agents That Can Reach Out First\n\nSentinel watches events, schedules checks, evaluates whether a nudge is useful,\nand routes proactive actions through guardrails. The point is not noisy\nnotifications; the point is an agent harness that can notice.\n\n### 5. Skills That Improve\n\nSkillForge treats skills as procedural memory. It can detect reusable workflows,\nwrite skill files, track execution feedback, and evolve instructions when they\nstop working.\n\n\u003cbr\u003e\n\u003cdiv align=\"right\"\u003e\n\n[![](https://img.shields.io/badge/-Back_to_top-gray?style=flat-square)](#readme-top)\n\n\u003c/div\u003e\n\n## Agent Templates\n\nRaven is an Apache-2.0 licensed, self-improving agent harness built by EverMind.\nIt provides the runtime, memory layer, tools, and Agent Templates for building\ncustom agents and digital workers.\n\nUse an Agent Template when you want Raven's harness layer but your own\nscenario, personality, workflow policy, skills, integrations, or distribution\nmodel. A template can start as one person's agent and later become a repeatable\ndigital worker for a team or community.\n\nAgents, templates, skills, workflows, and modules created with Raven belong to\ntheir creators. Builders may use, modify, commercialize, and share agents built\nwith Raven or based on Raven Agent Templates under the Apache-2.0 license.\n\nWe encourage builders to say \"Built with Raven\" and link back to this\nrepository. The Raven and EverMind names and logos may not be used to imply\nofficial endorsement unless explicitly approved by EverMind.\n\n## Useful Commands\n\n| Goal | Command |\n| --- | --- |\n| Start the native TUI | `raven` or `raven tui` |\n| Check the TUI runtime | `raven tui --check` |\n| Configure Raven | `raven onboard` |\n| Run a one-shot shell task | `raven agent -m \"...\"` |\n| Review providers | `raven provider list` |\n| List messaging channels | `raven channels list` |\n| Start the messaging gateway | `raven gateway` |\n| Manage sessions | `raven sessions list` |\n| Inspect scheduled jobs | `raven cron list` |\n| Browse skills | `raven skill list` |\n| Inspect proactive state | `raven sentinel status` |\n| Show plugins and memory backend | `raven plugins` |\n| Debug sandbox VMs | `raven sandbox list` |\n| Show local status | `raven status` |\n| Diagnose setup | `raven doctor` |\n\n## Docs by Goal\n\n| Goal | Start here |\n| --- | --- |\n| First-time install and setup | [Quick Install](#quick-install) |\n| Source-based development | [Developer Workflow](#developer-workflow) and [docs/dev.md](docs/dev.md) |\n| Memory and plugin architecture | [docs/memory-plugin-architecture.md](docs/memory-plugin-architecture.md) |\n| Sandbox usage and debugging | [docs/sandbox/usage.md](docs/sandbox/usage.md) |\n| Proactivity design | [docs/Proactivity-Plan.md](docs/Proactivity-Plan.md) |\n| Detailed design notes | [docs/README.md](docs/README.md) |\n\n\u003cbr\u003e\n\u003cdiv align=\"right\"\u003e\n\n[![](https://img.shields.io/badge/-Back_to_top-gray?style=flat-square)](#readme-top)\n\n\u003c/div\u003e\n\n## Architecture\n\nEvery turn flows through the Spine: one entry (`submit`), one exit (`emit`),\nand per-conversation lanes for ordering and cancellation. Feature engines plug\ninto the agent loop through explicit handoffs instead of importing each other.\n\n```text\nChannels / TUI / Gateway\n        |\n        v\n   Raven Spine\n submit -\u003e lanes -\u003e emit\n        |\n        v\n   Agent Loop\n tools · skills · providers\n        |\n        +--\u003e Context Engine   legacy / curator\n        +--\u003e Memory Engine    EverOS / local skills / SkillForge\n        +--\u003e Proactive Engine Sentinel / scheduler / nudge policy\n        +--\u003e TokenWise        usage tracking / cache placement / routing\n        +--\u003e Eval Engine      task judgement and coordination\n```\n\n### Repo Layout\n\n```text\nraven/\n├── spine/              # Per-turn backbone: submit -\u003e lanes -\u003e emit\n├── agent/              # Agent loop, tools, hooks, subagents, context builder\n├── channels/           # Telegram, Discord, Slack, Matrix, WhatsApp, WeCom, ...\n├── tui_rpc/            # Python side of the native TUI protocol\n├── providers/          # LLM provider adapters\n├── context_engine/     # Context assembly and Curator path\n├── proactive_engine/   # Sentinel, scheduler, nudges, feedback\n├── memory_engine/      # EverOS memory, local skills, SkillForge\n├── token_wise/         # Usage tracking, cache placement, routing\n├── sandbox/            # Isolated command execution\n├── security/           # Trust boundaries and network checks\n├── cli/                # `raven` command line entry point\n└── config/             # Config schema and update helpers\n\nui-tui/                 # React/Ink native terminal UI\nbridge/                 # WhatsApp TypeScript bridge\n```\n\n\u003cbr\u003e\n\u003cdiv align=\"right\"\u003e\n\n[![](https://img.shields.io/badge/-Back_to_top-gray?style=flat-square)](#readme-top)\n\n\u003c/div\u003e\n\n## Developer Workflow\n\nSource setup, focused checks, and PR rules live in\n[CONTRIBUTING.md](CONTRIBUTING.md) and [docs/dev.md](docs/dev.md).\nAI-collaboration rules live in [AGENTS.md](AGENTS.md); `CLAUDE.md` is kept as a\ncompatibility entry point.\n\n\u003cbr\u003e\n\u003cdiv align=\"right\"\u003e\n\n[![](https://img.shields.io/badge/-Back_to_top-gray?style=flat-square)](#readme-top)\n\n\u003c/div\u003e\n\n## Status\n\nRaven is pre-alpha and moving quickly. APIs can change without notice, but the\ncore product surfaces are already in the repository.\n\n| Layer | Status |\n| --- | --- |\n| Native TUI + CLI | Functional |\n| Spine runtime | Functional |\n| Base agent loop, tools, providers | Functional |\n| Context engine | Implemented, still evolving |\n| Sentinel proactivity | Implemented, still evolving |\n| TokenWise strategies | Implemented |\n| SkillForge | Implemented |\n| Eval engine | Partial |\n\n\u003cbr\u003e\n\u003cdiv align=\"right\"\u003e\n\n[![](https://img.shields.io/badge/-Back_to_top-gray?style=flat-square)](#readme-top)\n\n\u003c/div\u003e\n\n## Star Us\n\nIf Raven is the kind of agent harness you want to exist, star the repo. It\nhelps more builders of self-improving agents discover the project and gives the\nEverMind ecosystem a stronger signal to keep investing in open agents.\n\n### Star History\n\n[![Star History Chart](https://api.star-history.com/svg?repos=EverMind-AI/raven\u0026type=Date)](https://www.star-history.com/#EverMind-AI/raven\u0026Date)\n\n\u003cbr\u003e\n\u003cdiv align=\"right\"\u003e\n\n[![](https://img.shields.io/badge/-Back_to_top-gray?style=flat-square)](#readme-top)\n\n\u003c/div\u003e\n\n## EverMind Ecosystem\n\nEverMind is an open-source ecosystem for long-term memory, self-evolving\nagents, AI-native interfaces, and memory evaluation.\n\n\u003ctable\u003e\n\u003ctr\u003e\n\u003cth colspan=\"2\"\u003eEverMind Open-Source Ecosystem\u003c/th\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eSelf-Improving Agent Harness\u003c/strong\u003e\u003c/td\u003e\n\u003ctd\u003e\u003ca href=\"https://github.com/EverMind-AI/raven\"\u003eRaven\u003c/a\u003e - the terminal-native agent harness for tools, skills, memory, proactivity, context control, and reusable Agent Templates.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eMemory Runtime\u003c/strong\u003e\u003c/td\u003e\n\u003ctd\u003e\u003ca href=\"https://github.com/EverMind-AI/EverOS\"\u003eEverOS\u003c/a\u003e - the memory substrate Raven uses for durable user memory, agent memory, case extraction, skill extraction, and multimodal parsing.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eAlgorithm Engine\u003c/strong\u003e\u003c/td\u003e\n\u003ctd\u003e\u003ca href=\"https://github.com/EverMind-AI/EverAlgo\"\u003eEverAlgo\u003c/a\u003e - stateless extraction, ranking, parsing, and memory operators that power EverOS.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eHypergraph Memory\u003c/strong\u003e\u003c/td\u003e\n\u003ctd\u003e\u003ca href=\"https://github.com/EverMind-AI/HyperMem\"\u003eHyperMem\u003c/a\u003e - hypergraph memory for long-term conversations, with benchmark-backed topic -\u003e episode -\u003e fact retrieval.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eBenchmarks\u003c/strong\u003e\u003c/td\u003e\n\u003ctd\u003e\u003ca href=\"https://github.com/EverMind-AI/EverMemBench\"\u003eEverMemBench\u003c/a\u003e · \u003ca href=\"https://github.com/EverMind-AI/EvoAgentBench\"\u003eEvoAgentBench\u003c/a\u003e - evaluation suites for conversational memory and agent self-evolution.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eLong-Context Research\u003c/strong\u003e\u003c/td\u003e\n\u003ctd\u003e\u003ca href=\"https://github.com/EverMind-AI/MSA\"\u003eMSA\u003c/a\u003e - Memory Sparse Attention for scalable latent memory and 100M-token contexts.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003ePersonal Memory Layer\u003c/strong\u003e\u003c/td\u003e\n\u003ctd\u003e\u003ca href=\"https://github.com/EverMind-AI/EverMe\"\u003eEverMe\u003c/a\u003e - CLI and agent plugin suite for cross-device, cross-agent personal memory.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eDeveloper Integrations\u003c/strong\u003e\u003c/td\u003e\n\u003ctd\u003e\u003ca href=\"https://github.com/EverMind-AI/evermem-claude-code\"\u003eevermem-claude-code\u003c/a\u003e · \u003ca href=\"https://github.com/EverMind-AI/everos-plugins\"\u003eeveros-plugins\u003c/a\u003e - plugins, skills, and migration tooling for AI coding agents.\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/table\u003e\n\nTogether, these repositories form EverMind's research-to-runtime stack: memory\nmethods, reusable algorithms, benchmark evidence, native agent products, and\npractical developer integrations.\n\n\u003cbr\u003e\n\u003cdiv align=\"right\"\u003e\n\n[![](https://img.shields.io/badge/-Back_to_top-gray?style=flat-square)](#readme-top)\n\n\u003c/div\u003e\n\n## Contributing\n\nRaven is early, and useful contributions are welcome across runtime\narchitecture, TUI polish, provider support, memory workflows, proactivity,\nbenchmarks, documentation, and issue reports.\n\nBefore opening a PR:\n\n1. Read [AGENTS.md](AGENTS.md).\n2. Keep the change scoped.\n3. Add or update tests for behavior changes.\n4. Run the relevant `make` targets.\n5. Use a Conventional Commit title.\n\n### License\n\nRaven is licensed under the Apache License 2.0. Portions of the runtime and\nTUI layer originated from MIT-licensed upstream projects; their copyright\nnotices and license texts are retained in [NOTICES.md](NOTICES.md) and\n[LICENSES](LICENSES/).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fevermind-ai%2Fraven","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fevermind-ai%2Fraven","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fevermind-ai%2Fraven/lists"}