Zed retweeted
It’s time.
Alpenglow goes live on mainnet-beta Wednesday, October 21, ~14:50 UTC.
18 months of work for a migration that happens in one slot. Finality drops from seconds to ~150ms.
Make sure to follow us @anza_xyz as we countdown to go-live.
Details: anza.xyz/alpenglow
Zed retweeted
以前的 AI：帮科学家写写摘要、润色格式。
现在的 AI：从尘封快 30 年的废弃数据里，识别出人类当时看不懂的物理机制。
“人类的全部知识已经足够我们再突破几次技术革命，只是我们自己读不过来而已。”
你的下一篇顶刊，可能就在 90 年代某位博士生的废稿里。
原文：vals.ai/blogs/room-temperatu…
Zed retweeted
I've had agents implement and optimize the Campfire web app in Elixir, Go, and Rust. Yes, Ruby is the slowest. I didn't care about that when the pay-off was huge in terms of productivity and developer joy. But if you're no longer reading the code? github.com/basecamp/once-cam…
Readers added context they thought people might want to know
The shown benchmarks use non-equivalent implementations: Elixir routes all DB queries through a single GenServer with Redis for jobs (retained from Rails), while Rust uses in-process queues, omits CSRF tags/fields, and applies other caching changes.
github.com/basecamp/once-…
raw.githubusercontent.com/basecamp/once-…
github.com/basecamp/once-…
github.com/basecamp/once-…
x.com/zachsdaniel1/s…
Zed retweeted
Codex tip: once GPT-6.1 Sol runs your main session, stop paying Sol prices for the grunt work
hand the reading to Luna and put Astra on call
Sol keeps writing the code
Luna agents read the repo and pull the docs, fast and cheap
Astra only gets spawned at three moments:
→ before a plan: is this the right approach?
→ when the same error comes back: am I digging in the wrong place?
→ before "done": what did I miss?
Sol ships. Luna reads. Astra checks
jev engineering is the same move one layer down: the forks that need no thinker (which file, which tool, retry or stop) go to jev in under half a second, and the big models only see the ones that split
- the full tree
> GPT-6.1 Sol on high runs the main session
> explorer reads the code on Luna
> researcher pulls the docs on Luna
> worker edits and runs tests on Sol
> Astra on call as the architect
> auto_review watches every approval
paste the tree and this prompt into Codex ↓
"Rebuild my Codex setup around this tree:
1. Check ~/.codex/agents and .codex/agents for agents that already fit explorer, researcher and worker.
> Draft new TOML files only for missing roles
> explorer and researcher on gpt-6-luna, worker on gpt-6.1-sol, all with model_reasoning_effort medium
> Add an architect agent on gpt-6-astra, model_reasoning_effort high, that only reviews plans, repeated errors and finished work
> List any that pin a different model before changing them
2. In ~/.codex/config.toml set model to gpt-6.1-sol, model_reasoning_effort to high and approvals_reviewer to auto_review
3. Find anything that would override this (active profiles, flags in my shell aliases, agents.default_subagent_model). Report it, change nothing
4. Add one rule to AGENTS.md: spawn the architect before a large plan, when an error repeats, and before calling a long task done
Show me every change as a diff first. No edits until I say go."
↳ developers.openai.com/codex/…
Zed retweeted
Codex tip: once GPT-6.1 Sol is your main model, stop running Astra on every turn
put Astra on call as an architect agent
GPT-6.1 Sol keeps writing the code
Astra only gets spawned at three points:
→ before a plan: is this the right approach?
→ when the same error comes back: am I digging in the wrong place?
→ before "done": what did I miss?
Astra reviews. Sol ships
Jev engineering is the same move one layer down: the forks that need no thinker (which file, which tool, retry or stop) go to Jev in under half a second, and the big models only see the ones that split
- the full tree
> GPT-6.1 Sol on high runs the main session
> explorer reads the code on Luna
> worker edits and runs tests on Sol
> researcher pulls the docs on Luna
> all three on medium
> Astra on call as the architect
> auto_review checks every approval
paste the tree and this prompt into Codex ↓
"Rebuild my Codex setup around this tree:
1. Check ~/.codex/agents and .codex/agents for agents that already fit explorer, worker and researcher.
> Draft new TOML files only for missing roles
> explorer and researcher on gpt-6-luna, worker on gpt-6.1-sol, all with model_reasoning_effort medium
> Add an architect agent on gpt-6-astra, model_reasoning_effort high, whose only job is reviewing plans, repeated errors and finished work
> Skip any that pin a different model and list them
2. In ~/.codex/config.toml set model to gpt-6.1-sol, model_reasoning_effort to high and approvals_reviewer to auto_review
3. Find anything that would override this (active profiles, flags in my shell aliases, agents.default_subagent_model). Report it, change nothing
4. Add one rule to AGENTS.md: spawn the architect before a large plan, when an error repeats, and before calling a long task done
Show me every change as a diff first. No edits until I say go."
↳ developers.openai.com/codex/…
Zed retweeted
Cloudflare 团队把其内部找漏洞的那套方法做成了一个 Skill，叫 security-audit，并开源了。
装进编码 Agent 就能对自己的代码库跑一次正经的安全审计，已斩获 11000+ Star！
流程分 6 个阶段，先摸清架构和信任边界，再派出多个独立的 Agent 分片去找，每个候选漏洞交给一个新的 Agent 专门去推翻它。
GitHub：github.com/cloudflare/securi…
它的规矩是找到的那个 Agent 不能自己验，验证阶段的目标就是证明这条不成立，最后只有三种结论，确认、待核实、已排除。
待核实的会写清卡在哪个具体事实上，不给严重程度。少了一层防护但另一层挡得住的，只算加固建议，不算漏洞。
产出是一份主报告加一份逐条细节，每条确认的都能追溯到源码位置。
Cloudflare 博客说这就是他们内部漏洞发现系统的起点，多跑几次结果会叠加，他们自己测下来单跑一次大概只能找到一半。
用法就一句「security audit this codebase」，在自己有权限审的代码库上跑，把结果当一份第二意见看，我觉得比裸问 Agent 靠谱得多。
Zed retweeted
Wow
“Make Red Alert 2 (YR) compile and run natively on iOS & macOS.”
Codex (GPT-5.6 Sol) worked for 26 days, analyzed the EXE and rebuilt the entire game in 624K lines of C++
This is by far the hardest task I’ve ever given Codex as the game's code was never released and is believed to be lost.
>
Readers added context they thought people might want to know
There is already an open source effort to build this and the posts claim that it was decompiled and built without assistance is false as the information to unpack the assets and game mechanics and rendering is already in the training corpus github.com/hel1o-wor1d/Op…
Zed retweeted
SGP-002 can still be saved. While we can't convince Figment there are plenty of no/abstains to go after:
- Shinobi Systems - 1.4m
- All Nodes - 795k - @Allnodes
- Greed Academy - 546k @GreedAcademy
- Chainflow - 530k - @cfl0ws
- Phantom - 420k - @phantom
- Raiku - 356k - @raikucom
- Hanabi - 345k - @0x7cf
- Stakecraft - 276k - @stakecraft
- Blocksize - 250k
- Validator Open - 232k
- Solend - 220k (lol they are still on Solana?)
- Pine Stake - 218k @PineStake @JotaDotSol
- H20 nodes 189k - @H2oNodes @maxh2onodes
- MonkeDao 187k - @MonkeDAO
And lots more! -> perp.so/governance
@solana
Zed retweeted
imo Solana dApps shouldn’t be thinking about CU golfing yet and should continue to focus on PMF and attracting new users. We also need better forums to discuss these changes before they get sent to a vote. Glad the discussion sparked some debate but feels like poor timing
ORCA!!!
Great to see this come through. Been building toward this one for a while.
Every tokenized asset needs onchain liquidity - that’s the layer @orca_so is building, and this is the first of many.
Special thanks to @cpmontagano for being the closest through it all. Grateful to the Shinhan Asset Management team, @scottleesol, and @EtherfuseDave for working alongside us. The real work starts now.
Onwards and upwards!
Zed retweeted
100% faster slots, 90% cheaper rent, 400% transaction size increase
Solana v4.2
Activating soon.
Zed retweeted
For the most part, users don't care about decentralization.
They do, however, care about what decentralization provides in the long-term; lower fees, less-KYC, composability, ergonomics, frontier experimentation, lack of lock-in.
Therefore the goal as DeFi builders is to properly demonstrate that long-term value in the short-term. Unfortunately, this is not an easy task at all.
Centralized entities can probably hide their true nature for up to a year: they can lower fees in the short term, blur/delay KYC, open up composability and then rug it later. They also have capital today for short term incentives.
Eventually, however the chickens come home to roost. But before them; what do we do?
On one hand, we can target more educated users that have seen the same scenarios play out or have read about it. This, however, severely limits potential user-base.
Generally, the other option is probably incentives 'imported' from the future that a centralized entity cannot guarantee. There lies the big idea; the future.
The future is the realm of decentralization; it's where it is infinitely more valuable than centralization which has non-horrible characteristics now but always regresses to the mean. Decentralization, however, protects against the worst cases.
If you can open up a stable communication/value channel with the future; you can sell decentralization and you can align customers on a bigger vision.
It's why storytelling and vibes are so important in crypto; they can appeal to emotions which are more tapped into the future than thoughts.
Tap into the trifecta of education, storytelling and incentives.
Zed retweeted
profiling solana programs will never be the same
full symbolication through the entire call stack & across cpi boundaries
github.com/soundsonacid/seas…
