EverMind

DeepSeek-V4-Flash is already a first-class model in Raven 🐦‍⬛. Connect your DeepSeek API key and switch in one command: /model deepseek-v4-flash --provider deepseek Your tools, skills, working environment, and EverOS-backed memory stay with the agent. github.com/EverMind-AI/Ra…

Raven 🐦‍⬛ v0.1.10 is out. Choose from 21 providers, or search any provider LiteLLM can route to. Recognized deletion commands now ask for one-shot approval in the TUI. Each model call becomes a collapsible episode. Already using Raven? `raven upgrade` github.com/EverMind-AI/Ra…

Our new paper. Self-evolving agents have a measurement problem. The hard part isn’t generating changes to an agent harness. It’s knowing which changes genuinely helped rather than got lucky, overfit, or never activated at all. LLMs diagnose failures and propose changes across prompts, knowledge, runtime, tools, and configuration. But deterministic code owns the credit: validity checks, activation checks, paired significance testing, and a sealed test. Across 7 domains, the evolved harnesses gained +9.0 to +15.5pp on the 6 statistically credited sealed tests, retaining 86–147% of the training gain. What transfers is not one universal harness. It’s the diagnose-and-credit loop. arxiv.org/abs/2607.13683

EverOS 1.2.1 is out. LLM-only startup. Add embeddings and reranking later. Three runtime tiers + everos cascade backfill. Also: bounded cascade retries and a security fix for knowledge uploads. On 1.2.0? Upgrade now: `pip install --upgrade everos` github.com/EverMind-AI/Ev…

Raven 🐦‍⬛ just crossed 3,000 stars on GitHub. Thank you to everyone who has built, tested, and shared it with us. To celebrate, you can now add this tiny Raven to Codex as your pet. github.com/EverMind-AI/Ra…

@karry_viber @raft_hq 🤝👀

研究了一晚上Orb迁移Raft的事，发现之前自己在意的多agent失去主干“记忆”的事被我前阵子引入的everos解了！ 你们真不考虑合作下吗？@raft_hq @evermind

We’re proud to sponsor Archify. An open-source, MIT-licensed agent skill for creating beautiful, verifiable architecture, workflow, sequence, data-flow, and lifecycle diagrams, delivered as self-contained HTML with motion and crisp exports. Built by @t20000622yy, a member of the EverMind community. This is exactly what we hope to see more of: ambitious projects emerging from the community and built in the open. Fantastic work. Very proud. 💛 github.com/tt-a1i/archify

@OpenDesignHQ @Yangtze_Seventh What a collaboration!

Complex ideas deserve clear, memorable storytelling ✨ @Yangtze_Seventh used Open Design as part of the creative workflow behind this sticker-style walkthrough of the @evermind stack. Love seeing what creators are building with Open Design!

Raven TUI looks good in the light. Raven TUI looks good in the dark. Pick your theme. Go with Raven 🐦‍⬛.

A question we often hear: How are EverOS, EverMe, and Raven related - and what does each one do? Here’s a useful two-minute walkthrough from our DevRel @Yangtze_Seventh Sharing his take on the EverMind stack.

A skill library shouldn’t just grow. It should get better. EverMind Skill Hub improves 1,240 skills every week using real usage signals and SkillBench regression tests. Open. Measured. Editable. Self-evolving.

Yellow recognizes yellow. We believe in staying open too.

This shade has a name now: EverMind yellow. New banner unlocked. Hope you like it.

Welcome to X, Jensen! The future needs both open and closed models. It also needs memory that isn’t locked inside either.

@yaosiscom @yinmin1987 @elliotchen100 咱们做长期记忆呢 😀

@yinmin1987 @elliotchen100 @evermind evermind是啥 你被包养了🥹 尹佬 为啥不带上我一起

EverMind礼物收到啦 好棒👍 @elliotchen100 @evermind

Meet EverMind Skill Hub, a measured capability layer for AI agents. We aggregated 800K+ raw SKILL.md files and curated them into 96,401 license-audited, OSI-compliant skills. Then we built purpose-trained matching and tested the result across four real-world benchmarks, delivering statistically significant gains on every one. Search by task, inspect the source and quality signals, then install or adapt your own version. In Raven, the loop continues: EverOS preserves what works, while SkillForge turns repeated workflows and feedback into evolving local skills. Agents don’t need another directory full of skills. They need the right capability, at the right moment, with evidence that it works. evermind.ai/skillhub

Open source at its best. Yesterday, we opened an issue asking @OpenDesignHQ to support Raven as an MCP installer target. Within an hour, a community contributor picked it up. A few hours later, the PR was ready, and today, it was merged. Raven is now an officially supported MCP target in Open Design: `od mcp install raven` Huge thanks to roian6 and the Open Design community. This is how open source should move. 🐦‍⬛🤝🎨 Raven： github.com/EverMind-AI/ra… Open Design： github.com/nexu-io/open-d…