hiromi maeo

これは良さそうなアイデア。 ちょっと分析してみる。

🚨How do you index the entire Linux kernel (28M lines of code) for an AI agent in 3 minutes? You stop letting the agent read files one by one. There is a fascinating new open-source release called codebase-memory-mcp. It's a code intelligence engine that swaps traditional file-searching for high-speed AST knowledge graphs. What makes this project stand out is the research behind it. Evaluated across 31 real-world repositories (detailed in arXiv:2603.27277), the architectural shift yields massive efficiency gains: → 99% reduction in tokens for structural queries → 83% answer quality across complex tasks → 2.1x fewer tool calls required It maps functions, classes, HTTP routes, and cross-service links into a graph. When the agent needs context, it queries the graph directly. Security is prioritized too: everything happens 100% locally on your machine via a single static binary. It runs entirely locally. No Docker, no Ollama, no API keys. You download the binary, restart your agent, and it just works. Are we one good index away from cutting AI dev costs to zero? Paper and Repo links in the thread ↓

ビズリーチのCTOの講演資料を読んで驚いた。 月間2,300億トークンをCodexで消費する組織(350人)が辿り着いた結論が、私が自分のために組んできた仕組みとほぼ同じ構造だったから。 講演の核はシンプル。 — harnessは借りられる。統治は、自社で作るしかない。 harness=AIが正しく速く安全に動く足場(実行環境・文脈・ガードレール・検証ループ・観測性)。これはAnthropicやOpenAIから借りられる。でもそれだけだと「人間がフローの中にいるまま速くなる」=HITLの高速化止まり。人間が律速のまま抜け出せない。 だから、ここから抜け出すには統治がいる。 そこで彼らの出した答えが三権分立だった。 ●立法=何が正しいかを定義する(ルールのSSOT) ●司法=従ってるか裁く(機械チェック+LLM意味判定) ●行政=実際に実行する(workflow/agent) そして全部を、人間しか改正できない憲法が縛る。 で、改めて自分のシステムを見たらほぼ全部あった。 ルール群=立法、Phase 5.5のゲート=司法、複数体のagent=行政、Design Constitution=憲法。 司法を「Grep/ASTの決定論チェック+LLMの意味判定」に二層化してるところまで一致してて、これは正直驚いた。 特に「書かれてるだけ」を許さない構造に関して。 ADRやWikiに書いてもagentは参照しないし、違反しても誰も止めない。だから、うちの環境ではCIが落ちる・hookがEdit自体を止める仕組み。 そして、「書かれている」と「効いている」は別物 — これが資料で一番刺さったところ。 === ここまでが一致の話。ここからはうちの環境の穴。 資料には2つのグラフが出てくる。ルールとチェックを双方向に繋ぐAuthority Provenance Graphと、機能→仕様→テストを追うSpecification Provenance Graph。うちの環境にはこれが無い。 つまり効かせるゲートは持ってるのに、ルール↔チェック↔実行を横断で追跡する地図が無いんよな。 だから「チェックはあるが何のルールに依拠してるか不明」みたいな断絶を、機械的には検知できない。 統治は、止められるだけじゃ足りない。繋がりを追跡できて初めて完成するんじゃないか、と。 とりあえず次に埋めるのはそこ。

New update on Fable 5: and it's less about jailbreaks than anyone initially thought. Via Axios The Axios story that just dropped today reframes the whole thing: Anthropic hired a cybersecurity expert to review Amazon's findings and push back on the government's narrative. The administration viewed her as a "radical Democrat." She was then publicly celebrated by Chris Krebs, the official Trump just fired. That didn't help. Behind the scenes, officials describe a company that simply doesn't know how to talk to this administration. "It's like they just speak different languages," one source said. "Everybody said Anthropic was a bad actor. Some of us said it was time to give them a chance. Now those people are questioning that. They screwed us." Today: Anthropic staffers meet with Commerce, the CIA, and White House science advisor Michael Kratsios to work through compliance with the cyber executive order. The technical question - can Fable 5 be jailbroken - is almost secondary now. This is a story about a company that keeps losing the room. Ill keep you updated.

Researchers show that Claude Code is 98% not AI. Anthropic never gave us the architecture for Claude Code. There were no docs. Just a tool that every developer is currently obsessing over. Until it leaked recently. A research team pulled the source code, analyzed all 500,000 lines, and found something ridiculous. Only 1.6% of the codebase actually interacts with the AI model. The core of Claude Code is literally just a simple while-loop. It asks the model what to do, runs a tool, and repeats. So what is the other 98.4%? It is hardcore, traditional software engineering. The researchers found a massive, complex infrastructure designed entirely to babysit the AI and keep it from hallucinating or destroying your computer: - A 7-mode permission system acting as a security bouncer. - A 5-layer context compaction pipeline so the AI doesn't forget its goal. - A subagent delegation mechanism with strict worktree isolation. - Four different extensibility hooks to manage external tools safely. Every startup right now is trying to build a better AI model to get better results. Anthropic did the exact opposite. They took an existing model and built a fortress of deterministic software around it. They realized that the AI doesn't need to be smarter. It needs to be managed.