hello, world! i'm daniel! currently i'm the head of r&d of MiniMax code @MiniMax_AI, which is a frontier ai lab in china.
i enjoy working on our ai-native product @MiniMaxAgent . i'm really obsessed with this question: what does an ai-native organization look like? how can our organization co-evolve together with llms?
besides that, i'm passionate about open source. i'm the co-author of the deerflow project, which is a representative work in harness engineering and has 78k 🌟 on github.
if you're passionate about open source, foundation models, or agents- or if you enjoy playing basketball or riding a road bike - let's chat and be friends! :)
I put Anthropic's whole harness playbook for long-running coding agents on one page
honestly the most useful thing I've read on agents all year
two Anthropic engineering posts + the Opus 5.5 launch, squeezed into one sheet:
> a planner grows a one-line idea into a 16-feature spec
> a generator builds it, an evaluator opens the app in Playwright and fails any sprint below the bar
> session one writes https://t.co/K47G5WnxJt, a progress log and 200+ failing features. every session after closes one and commits
> the 4 ways agents wreck long tasks, and the fix for each
> 3 prompts to copy: coding agent, QA agent, planner
solo run: 20 minutes, $9, the game didn't even respond to input
full harness: 6 hours, $200, playable
QA was about 8% of the bill on their DAW run
and Opus 5.5 runs about 40% cheaper than Opus 5
bookmark this before you wire up your next agent
@Cyrbuzz123 That's exactly why it's a published standard, not internal glue. The contract is what apps build against, and we'll treat breaking changes to it carefully.
Mini Apps are now an open standard in MiniMax Code: a port, an Agent, a data directory. You describe the tool, the Agent builds it.
https://t.co/u1ZJdNbg32
Built something useful? Contribute it 👇
https://t.co/X1XuP404GY https://t.co/CejbIMn4xk
@AlgoritmaTurk@MiniMax_AI@MiniMaxAgent Thanks for digging in and confirming the fix! The MCP transport timeout with no reconnect is a great catch. Mind opening a separate issue for it so we can track it on its own? 🙏
@aiclawdius You do, with the Agent. That's the point: the person who uses the tool can change it in a sentence, instead of waiting on someone who owns it. And if it's useful to others, PR it to the community repo and it gets maintained there.
@swordBluesy Agreed, static declaration is step one. Runtime checks on child processes and permission changes are on our list. Thanks for pushing on this.
@djekyll Partly enforced: when a page hands something to the Agent, it lands in your input box first, and tool calls go through approvals. The rest is how you design your app.
@11Shixi Fair point, and you read the docs closely 🙏 Before the old process stops, the host can roll back. After that, a failed swap should say so clearly. Noted for the UI.
@Rosasynthesiz It's an open standard on top of the MiniMax Plugin protocol, and the community repo is MIT. Today it runs in MiniMax Code. If other harnesses want to host it, we'd love that.
@PinakCorp Tool approvals show up right inside the page, so you see what the app is about to do before it happens. And /miniapp-creator starts with read-only queries against each data source before it writes anything.