Released Oct 23, adding a tool with type: CODING_AGENT to a CREATE MCP SERVER spec gives any MCP client an agent that writes and runs code in a managed sandbox attached to the conversation thread. Clients send a text argument and pass thread_id to keep going in the same sandbox (example).
Shown on HN Oct 26, the AGPL 3.0 beta stores every conversation turn verbatim on your own PC with secrets masked, then hands any MCP connected agent a 3k token packet of the turns that answer a question about the past; the author reports 88.8% on LongMemEval. Install with pip install graph mind memory && graph mind install , or claude mcp add graph mind memory python m graph mind.mcp server for tools only; it's used daily on Windows, while macOS and Linux only pass CI so far…
Released Oct 26, the extension lets you pick a separate model for cloud tasks under Settings API Configuration, shows a progress card while a cloud sandbox clones your repo, and makes GPT 6 models work on OpenAI and Azure compatible endpoints. With no model chosen, the Cline provider goes back to Claude Sonnet 5.5, and most other providers now default to Kimi K3.1 Pro, MiniMax M3.1 Flash or DeepSeek V4.1 Flash, so pin your model if you don't want a surprise switch…
Released Oct 26, a compaction size estimate that ran about 7x too high (triggering compaction every few tool calls) is fixed, and the gemini provider no longer sends native Google Search, which made Gemini 3+ free tier keys fail every turn with 429s. Starting a session mid run now reattaches to the live run instead of orphaning it, model retries drop from 5 to 3, and ClinePass now defaults to Kimi K3.1 Pro…
Shown on HN Oct 26, the AGPL 3.0 Go binary starts a hub and agent host with kolo up and prints an invite link, so teammates can watch, type prompts, approve steps or take over the same terminal without installing anything, while the host keeps read only seats for observers. It's an experimental alpha where agents run with the host's permissions, so the README says to start in a dedicated VM (Show HN)…
Shown on HN Oct 26, the tool runs a shared counting semaphore so OpenCode, Codex CLI, Claude Code and other agents across repos or worktrees take turns on heavy jobs like cargo test while queued agents keep coding, with permits freed when the owning process exits. Install with npm install g @aron98/load lock then load lock install harness claude code (or codex, opencode, cursor); it's Linux first and the Claude Code and Cursor hooks aren't live verified yet…
Spotted something we missed? Start a thread.