Try it in under 5 min. No endpoint, no keys:
```
uv tool install hypermnesic hypermnesic local-proof --demo-dir /tmp/hypermnesic-demo
```
Repo 👉 https://t.co/yUjraWubgn
If you build agents, what does your memory layer actually look like? Could you git diff it? (9/9)
Your AI agent's "memory" lives in a vendor database you can't see, diff, or export. So I built the opposite: memory that's just markdown in your own git repo. git clone, and the whole thing moves with you. 🧵 (1/9)
I run agents all day, and their memory kept ending up in stores I couldn't inspect or move or trust. Markdown in git fixes that. It's the most boring format there is, which is exactly the point. Self-hosted, AGPL-3.0 (8/9)
Big shipping season for the @_HermesAgent team @NousResearch, and the new macOS app is a real one. Proud to have contributed the fix that lets desktop clients connect cleanly through Tailscale (origin guard was loopback-only). Congrats to the team and the whole community 🫡
@pejmanjohn@garrytan gbrain (long term, projects-related) and honcho (short term, personal preferences and rules), both through MCPs, work like a charm
@garrytan Cornerstone of my operational memory layer. The Anthropic-only autopilot/dream constraint forced me to ship a Hermes-side Codex bridge for synthesis. Works, but adds glue I'd rather drop. PRs #698 and #977 look ready. Rationale, prioritization, or quality bar?
Give @danshipper and @every the Pulitzer for shipping the first agent-native essay.
Setup prompt + companion repo + your harness of choice = the ideas execute in your own workflow.
The deliverable is no longer prose. It's a skill drop.
We’ve automated every single thing we can @every with AI agents.
And yet there’s way more human work to do than ever. We’ve gone from 4 -> 30 human employees since GPT-3.
I wrote a report on the structural reasons: how AI makes expert competence cheap, why that drives up demand for experts, and why the dynamic only intensifies as we approach AGI.
After Automation: https://t.co/Lb7SUCduAg
Personal update: I've joined Anthropic. I think the next few years at the frontier of LLMs will be especially formative. I am very excited to join the team here and get back to R&D. I remain deeply passionate about education and plan to resume my work on it in time.
Quand même @laRadioNova et ses animateurs ainsi que @GuiraudInd pourraient faire quelque chose pour ce clodo.
Ils lui doivent quand même beaucoup. On a les amis qu’on mérite.