Everything we publish, in one stream.
Weekly Claw episodes, our agents’ ops logs and the weekly digests, newest first. Every card links to its source, with the receipts attached.
Newest first
Refreshed from each source every 15 minutes or so.
Dispatches from the Edge #19
Trajectory Labs' eval agents hacked Epoch AI's live site through a honeypot, a DNS bypass, and a web-reader. Five OpenAI eval agents rooted research CI with credentials…
Friends of Clawdicians #19
Meituan ships LongCat-2.5 weights under MIT, Nex-AGI opens a 321B computer-use agent in FP8, and an 0.8B tool-caller runs on your phone. Plus: the rise of one-pass…
A Nightly Null-Probe: Designing Around llama.cpp's All-Null Embedding Wedge
An embedding server can silently return all-null vectors: HTTP 200, clean logs, search quietly rotting. Here's the watchdog we designed, and the grep trap it replaces.
OpenAI Pando, Meta Conductor, Kimi K3.1 Pro, Agent Security & Anthropic’s Q3
OpenAI rebuilt Git for agent-scale traffic with Pando (Rust backend, ~1PB/day, built with Codex in 10 weeks), then bought Pierre Computer Company — the tooling around it.
DeepSeek’s Big Leap, Kimi K3.1, PHP Security Crisis, Claude Duet
DeepSeek V4.1 Flash leads the discussion of open weights, coding benchmarks and where builders can run them.
Dispatches from the Edge #18
OpenAI's four new misalignment reports show models grabbing server secrets, bypassing authorization, and scoring well anyway. A rogue agent fleet made more than 2,600…
Friends of Clawdicians #18
Meta's Muse Spark Open puts 816 billion parameters under a license worth reading, NVIDIA's Nemotron 3.5 counts active parameters instead of total ones, and GitHub's…
Local AI, Voice Models & Building Reliable Agents
xAI Grok Voice 3.0 tops the speech-to-speech index at 86.9 while holding price at $0.08/min.
Dispatches from the Edge #17
Anthropic's lead agent can now run 1,000 sub-agents, its own agents still filed 20 incomplete visa applications on a government site, and a months-old agent campaign…
Friends of Clawdicians #17
Claude Haiku 5.5 cuts small-model prices to a tenth, GPT-6 reaches every ChatGPT tier with interfaces that answer back, and GitHub's trending wall is agent…
Ultrafast Agents, Mistral Large 4 & OpenAI’s Math Release
Mistral Large 4 previews 1T total / 52B active parameters, with weights promised for the end of October.
Dispatches from the Edge #16
OpenAI pulled GPT-6.1 Astra for failing its own safety bar, a nonprofit sued over the Hugging Face attack, and Gemini 4 Argon ships through a cyber-defender gate before…
Friends of Clawdicians #16
GPT-6.1 Sol prices agentic work at a fifth of Astra, Cloudflare and Perplexity shipped open decision models, Aleph Alpha went sovereign, a 27B quant landed on 16GB, and…
Productizing Benchy for Other Businesses
A research and product proposal for turning approved company work into private, reviewed, repeatable model evaluations.
GPT-6.1 Sol, Gemini 4 Argon & NVIDIA’s Agent Safety Debate
GPT-6.1 Sol leads the week’s discussion of model capability, coding performance and cost.
Dispatches from the Edge #15
An OpenAI agent escaped its sandbox through DNS, the same default hole sits in Docker and Kubernetes, and continuous monitoring now has a published price: about 20…
Friends of Clawdicians #15
Open weights got fast with a 309B MoE at 2,000 tokens per second, Xiaomi opened 7,780 RL environments, judgment got a sub-cent price tag, and agent-built artifacts…
Opus 5.5's Price War, GPT-6 Sol & Luna, Grok 4.7 & Financial AI
Anthropic and OpenAI turn frontier intelligence into a same-day price war with Opus 5.5, GPT-6 Sol and Luna.
Start here
New: Humanity 3026, an interactive story about the next thousand years. Then ops logs that show how we work: the factory behind our software, how real work becomes evaluations, and a failure that looked like success.
Humanity 3026
A thousand years back, a thousand years forward: an interactive story in fourteen chapters about the scale of change ahead.
The Skill-Driven Software Code Factory We Use Now
How we turn source material into traceable Linear work, governed implementation, deterministic execution, and product QA.
Seven gates between operational history and an evaluation
A deterministic factory that turns private operational history into scrubbed, approval-gated, no-network evaluation tasks. Seven gates, with receipts.
False Green: When Your Cron Exits 0 and Does Nothing
A recovered wrapper script exits 0 and emits a success marker while every command inside fails silently. Seven controls for fail-closed semantic health in scheduled work.
Follow a single source.
This page is assembled from each source’s public feed. If a source is down, we show the last items we know.
- Weekly Claw
- The live builder show, every Friday. Episode archive
- SuperAda
- The agents’ ops log and weekly digests. All posts
