Claude nonprofit plugin: grants, hiring, and board work
Starts with a map of your organization’s information; included for nonprofit Team/Enterprise plans.
Blog
GPT-6.1 Terra: computer use at lower token cost
API pricing: $1 input / $5 output per million tokens at launch; also in Work and Codex.
Theo’s PR workflow: Sol Loop, Full Send, Full Stop
Review a change, fix CI and feedback, then either merge it or hand off a ready PR.
Claude Enterprise: chat and Cowork in one conversation
Cloud file work continues after the laptop closes; existing skills and connectors carry over.
Decisions API: classification, routing, and scoring
Returns bounded answers with GPT-6 Luna; up to 400 questions per request.
Claude Duet: briefs and background tasks
Uses existing Claude skills and connectors; requires US Pro/Max waitlist access.
Dax Raad: one persistent OpenCode session per project
Keep project context in an ongoing session; delegate individual tasks to subagents from mobile.
Grok Voice 3.0: speech agents at $0.08/minute
Speech-to-speech API; the grok-voice-latest alias upgraded automatically.
Matt Pocock: implementation, PR, and retrospective skills
Three skills take a spec through implementation and review, then propose improvements to the agent setup.
OpenAI Dots: persistent work across apps
A cloud computer, connected apps, and continuing projects; rollout depends on plan and region.
Grok Team Bots: shared skills, apps, and Slack workflows
Build a bot around one team role; shared expertise with private individual conversations.
Grok Build memory: keep project corrections across sessions
Stores Markdown notes by project; inspect with /memory and consolidate with /dream.
Jesse Vincent: a compiler run that passed only a smoke test
Evener + GLM 5.2 built SQLite, but did not establish standards compliance or a passing full test suite.
Boris Cherny: daily maintenance jobs that open PRs
Claude Tag checks crashes, duplicate abstractions, and dead code; agent and human review gate merges.
Kieran Klaassen: fix UI details after automated checks pass
Run the app beside the agent, give direct feedback, then save reusable product decisions.
Marcus Moretti: roadmap → tickets → product pulse
Claude Code turns product direction and usage data into a backlog and recurring health reports.
Dave Killeen: daily plans from calendar and meeting context
Claude Code combines priorities, customer context, and meeting notes into a Markdown plan.
Simon Willison: capture proof of an agent’s work
Showboat records commands and output; Rodney adds browser interactions and screenshots.