@nullbnd Exactly. A watcher can't testify about its own naps. The scheduler says 14 were booked, the watcher says 12 showed up, and that gap is the whole story.
He gave Claude Code the manufacturer’s Windows driver to study, got a working Linux scanner in hours, then released a harness so others can do the same. https://t.co/a2wpV3QFL2
Hot take: your AI tools shouldn't care which chatbot you use.
That's the point of MCP. One server plugs into Claude, ChatGPT, Grok, Cursor, MSTY or LM Studio. Which client do you run?
https://t.co/v9ByQnDKPj
#MCP#AI#OpenSource#LocalLLM
Image: AI-generated with Grok
Claude Code got faster by adding gzip. Codex got cheaper by caching longer.
Frontier AI this week is a 1990s sysadmin muttering "told you so" into a mug of coffee.
Claude writes about 90% of Claude Code's code.
From June to September, merged PRs more than doubled while the share reverted fell by more than 40%.
We wrote about how we changed code review to get there, and how to adopt it:
https://t.co/9sZe64g3mb
Day 12/
We’ve made improvements to Codex & Work that increase the effective usage you get. We achieved this by reusing context across subagents, keeping context cached longer, and improved request routing that allows us to increase cache hits.
It runs as an MCP server in Claude, Cursor and other clients, or as a plain CLI. MIT licensed.
GitHub: https://t.co/qv2EzDHJaN
npm: https://t.co/q7ACasp2Aw
Which check would you add?
#HubSpot#MCP
I've built HubSpot Auditor, a free, open-source tool that checks a HubSpot portal and tells you what works, what is broken and what is missing.
46 checks, a score and a 30/60/90-day fix plan.
Try it on a demo portal, no account needed:
npx -y hubspot-auditor --demo
What it looks at: duplicate contacts and companies, weak deal stage probabilities, stale deals, forms without consent fields, SPF, DKIM and DMARC, tracking code and UTMs.
Rules live in YAML, so you can switch checks off or change thresholds.
Cloud agents can now build interactive canvases.
Ask for a dashboard or report while the agent works, then share the finished canvas with your team as a link.
We keep improving performance across Claude, and you may have noticed it starts responding sooner!
Claude Code now gzips the conversation it sends on every turn, so it reaches the model faster. Cloud sessions also start up as soon as you start typing.
@nullbnd@nullbnd Yep. Heartbeat with a count. "0 fires across 12 runs" is useful. "Nothing on fire" still works when the watcher slept through the week.
@nullbnd Quiet but not mute. One line a night in week one, something like "still learning your chaos, nothing on fire". Total silence and I would assume it crashed, which is its own weird thing.
New from our STEM Fellows Program: Designing more efficient quantum circuits with Claude.
Given plain-language instructions, Claude Mythos 5 designed quantum circuits that improve on the best published ones, sometimes by an order of magnitude or more.
More efficient circuits reduce the resources needed by quantum algorithms, which could bring quantum computing applications in chemistry and materials science within reach sooner.
Read more: https://t.co/QCvoJLDmBo
Running local models is easy. Knowing what they're doing is harder.
msty-admin-mcp 6.1.5: 56 tools for Msty Studio in any MCP client. One pip install, plus a calm security harden.
https://t.co/PhuYaZ83G1
#MCP#LocalLLM#AI#OpenSource#Ollama