Fetching from the wire…
01
02
03
04
05
06
07
08
Tools2026-09-03 · source-backed
Released September 3, listed as a bug fix, with an explicit "Allow Self-Signed Certificates" toggle for people who need the old behavior. Which means every WebDAV backup this desktop client made before 2.0.11 went out without certificate validation. Same release adds a per-assistant or global context compression threshold and one-hour Anthropic prompt cache retention configurable from API settings.
Each link below shares sources, entities, or timing with this story.
Cherry Studio ships unified access to OpenAI, Anthropic, Gemini, DeepSeek, Qwen, Ollama, and dozens more providers in a single Electron app. Autonomous agent mode, built-in knowledge base, MCP support. It's basically a free, local-first alternative to switching between web int...
CherryHQ/cherry-studio (50,068 stars) cut v2.0.0 on August 5 and reached 2.0.2 by August 7, rebuilding from chat client into a work surface with multi-window and split-view layouts. The release notes say the subscription part explicitly: OpenAI and Anthropic endpoints, reusing...
perf(agent): stabilize system prompt prefix (#19297) keeps the prompt cache warm across turns rather than invalidating it whenever the prefix shifts. Also parallelized independent config pipelines, parallelized device metadata reads during backup, and anchored relative dates i...
Tools, connections, skills, instructions, and hooks can now be packaged into a single unit, published to package registries, then installed and versioned like any dependency (Vercel). This is the versioning and reuse story that ad-hoc skill directories have been missing, and i...
A solo Claude Opus 4.5 agent spent $9 and 20 minutes building a retro game. It was broken. The same model, wrapped in Anthropic's multi-agent harness, spent $200 over 6 hours and produced a fully playable game with physics, sprite editors, and AI integration. Anthropic's engin...
Released September 3, v2.38.0 adds context_window to ModelProfile and context_window_used to RunContext (#4611), giving agent code a first-party way to read remaining context instead of estimating from token counts. It also adds a VLLMProvider for self-hosted vLLM servers, Cla...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.