The daily firehose — everything the toolchain shipped today, already filtered.
// HOW THIS ISSUE IS MADE
We read every release from the 174 tools on our watchlist at the source — GitHub and GitLab release notes, vendor release pages and changelogs, project blogs and feeds, vendor press releases, and the source code behind the tag. Bug-fix-only releases and non-product newsroom noise are dropped; what's left is summarized down to the new capability, how to try it, and any screenshots or videos the release itself published. Every entry links to the sources it was built from.
AutoGPT Platform v0.6.43 adds hybrid search with BM25 reranking, a chat copilot with agent generator tools, safe mode toggles, and a new LLM Picker UI.
└──▷ GET THIS VERSION
$ git clone --branch autogpt-platform-beta-v0.6.43 https://github.com/Significant-Gravitas/AutoGPT.git
# already have the repo? check out this version:$ git checkout autogpt-platform-beta-v0.6.43
›Adds explicit safe mode toggles for Human-in-the-Loop (HITL) and sensitive actions, giving operators direct control over when agent execution pauses for approval.
›Adds hybrid search across store listings, docs, and blocks — including embedding backfill for all content types and BM25 reranking for chat search.
›Adds a chat copilot with agent generator tools, enabling conversational agent creation from the builder interface.
›Adds a new LLM Picker UI for selecting language models directly in the platform.
›Adds a ConcatenateListsBlock for merging list outputs within agent graphs.
+4 moreshow less
›Adds a block documentation auto-generation system to keep block references current.
›Adds an interactive tutorial for the new builder interface to onboard new users.
›Adds an endedAt field to execution records for more accurate execution analytics timestamps.
›Improves Langfuse tracing with v3 SDK patterns and @observe decorators for better agent observability.
LangGraph prebuilt 1.0.7 adds dynamic tool calling via a tool override in wrap_model_call.
└──▷ GET THIS VERSION
$ git clone --branch prebuilt==1.0.7 https://github.com/langchain-ai/langgraph.git
# already have the repo? check out this version:$ git checkout prebuilt==1.0.7
›Supports dynamic tool calling by accepting a tool override parameter in wrap_model_call.
›Adds Lobster optional plugin tool for typed workflows and approval gates.
›Adds custom assistant identity and avatars configurable via the Control UI and agent config.
›Adds clawdbot update wizard with interactive channel selection, restart prompts, and preflight checks.
›Adds /models command and expands clawdbot models paging for improved model listing UX.
›Moves gateway service commands under clawdbot gateway, flattens node service commands under clawdbot node, and adds gateway probe for reachability checks.
+12 moreshow less
›Adds elevated exec approval modes (ask/full) with allowlist gating and approval tables rendered on write.
›Adds gateway/node targeting, wildcard agent allowlists, and target details to exec approvals.
›Adds per-channel idle duration configuration via sessions.channelIdleMinutes.
›Adds cache.ttlPrune mode with auth-aware defaults to reduce token spend on cold requests.
›Adds per-channel debounce overrides for auto-reply queue behavior.
›Adds wildcard channel config support for Discord.
›Adds typing indicators and DM read receipts to the Signal channel via signal-cli.
›Adds file uploads, adaptive cards, and attachment handling to the MSTeams channel.
›Adds explicit session keys and active hours support for Heartbeat.
›Adds cache trace config for debugging diagnostics.
›Adds macOS Settings refresh: location access in Permissions, connection mode in menu.
›Enables nodes to run exec-style and exposes PATH in node status/describe and bootstrap.
└──▷ BREAKING ON UPGRADE
!The Control UI now rejects insecure HTTP without device identity by default; use HTTPS (Tailscale Serve) or set gateway.controlUi.allowInsecureAuth: true to allow token-only auth.
!Envelope and system event timestamps now default to host-local time instead of UTC.
browser-use 0.11.4 ships a new CLI with headless/headed/real Chrome/remote browser modes and named multi-session support.
└──▷ GET THIS VERSION
$ git clone --branch 0.11.4 https://github.com/browser-use/browser-use.git
# already have the repo? check out this version:$ git checkout 0.11.4
└──▷ TRY IT
Automate a login flow against a local dev server using your real Chrome profile so saved credentials are already present.
$ browser-use --browser real open http://localhost:3000
browser-use state
browser-use click 0
browser-use type "mypassword"
browser-use screenshot login-result.png
browser-use close
Run parallel browser sessions to compare a work app and personal email side-by-side without sessions interfering.
$ browser-use -s work open https://work.example.com
browser-use -s personal open https://gmail.com
browser-use sessions
Use the cloud stealth browser with proxies for a site that blocks headless traffic, using your API key.
$ BROWSER_USE_API_KEY=<your-key> browser-use --browser remote open https://example.com
browser-use screenshot result.png
browser-use close
›Adds browser-use CLI with subcommands open, state, click, type, screenshot, and close for scripting and agentic browser control from the shell.
›Adds --headed flag to browser-use open for a visible browser window instead of the default headless mode.
›Adds --browser real global flag to browser-use open to drive the user's existing Chrome profile (with saved logins).
›Adds --browser remote global flag to route sessions through a cloud stealth browser with built-in proxies and anti-detection, requiring the BROWSER_USE_API_KEY environment variable from cloud.browser-use.com; returns a live preview URL.
›Adds -s <name> flag for named, parallel browser sessions (e.g. browser-use -s work open <url> and browser-use -s personal open <url> running concurrently).
+3 moreshow less
›Adds browser-use sessions subcommand to list all active sessions.
›Adds browser-use close --all to tear down every active session at once.
›Adds a Claude Code/Codex skill installable at ~/.claude/skills/browser-use/SKILL.md that teaches agents to use the new CLI for local and remote browsing.
Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.
Claude Code v2.1.16 adds task management with dependency tracking, native VSCode plugin management, and remote session browsing for OAuth users.
└──▷ GET THIS VERSION
$ git clone --branch v2.1.16 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:$ git checkout v2.1.16
›New task management system with dependency tracking between tasks.
›[VSCode] Native plugin management support added to the IDE extension.
›[VSCode] OAuth users can now browse and resume remote Claude sessions from the Sessions dialog.
Pi v0.49.3 adds inline bash expansion, AI image-gen extension, custom share viewer URLs, and a new word-delete hotkey.
└──▷ GET THIS VERSION
$ git clone --branch v0.49.3 https://github.com/earendil-works/pi.git
# already have the repo? check out this version:$ git checkout v0.49.3
└──▷ TRY IT
Run an inline shell command inside a prompt without leaving Pi — useful for injecting dynamic values like timestamps or file contents.
$ # In a Pi prompt, type:
Summarize this file: !{cat README.md}
›New markdown.codeBlockIndent setting to customize code block indentation in rendered output.
›New inline-bash.ts example extension that expands !{command} patterns inline within prompts.
›New antigravity-image-gen.ts example extension for AI image generation via Google Antigravity.
›New PI_SHARE_VIEWER_URL environment variable to point Pi at a custom share viewer URL.
›Adds Alt+Delete hotkey for deleting a word forwards in the editor.
└──▷ BREAKING ON UPGRADE
!The tree selector label filter shortcut is changed from l to Shift+L; any workflow or muscle-memory relying on l to filter labels will now need to use Shift+L.
Qwen Code v0.8.0-preview.1 adds /bug in non-interactive mode, a redesigned welcome screen, and YOLO-mode visual cues.
└──▷ GET THIS VERSION
$ git clone --branch v0.8.0-preview.1 https://github.com/QwenLM/qwen-code.git
# already have the repo? check out this version:$ git checkout v0.8.0-preview.1
›Adds /bug command support to non-interactive mode, enabling automated bug-report submissions without an interactive session.
›Redesigns the CLI welcome screen and settings dialog with an updated layout.
›Uses dim border colors to visually distinguish YOLO/auto-accept mode from normal operation in the CLI.
Qwen Code v0.8.0-nightly adds /bug in non-interactive mode and redesigns the CLI welcome screen and settings dialog.
└──▷ GET THIS VERSION
$ git clone --branch v0.8.0-nightly.20260122.21b26a40 https://github.com/QwenLM/qwen-code.git
# already have the repo? check out this version:$ git checkout v0.8.0-nightly.20260122.21b26a40
›Adds /bug command support to non-interactive mode, enabling automated bug reporting from scripts and pipelines.
›Redesigns the CLI welcome screen and settings dialog with a new visual layout.
›Uses dim colors for YOLO/auto-accept mode borders, making unattended-approval sessions visually distinct.
llama.cpp server now preserves reasoning_content in assistant messages, enabling round-trip chain-of-thought workflows.
└──▷ GET THIS VERSION
$ git clone --branch b7813 https://github.com/ggml-org/llama.cpp.git
# already have the repo? check out this version:$ git checkout b7813
›Adds reasoning_content input support to the server, allowing assistant messages that include reasoning content to be passed back in subsequent requests.
llama.cpp b7793 adds partial OpenAI Responses API support at POST /v1/responses including tool calls and image/file inputs.
└──▷ GET THIS VERSION
$ git clone --branch b7793 https://github.com/ggml-org/llama.cpp.git
# already have the repo? check out this version:$ git checkout b7793
›Adds POST /v1/responses endpoint to the llama.cpp server with partial OpenAI Responses API compatibility, supporting text, image, and file input message types.
›Adds tool call support (function calling) to the /v1/responses endpoint for both streaming and non-streaming modes.
›Feeds reasoning/thinking model outputs through the chat template pipeline in the /v1/responses endpoint.
›Rejects input_file input type on /v1/responses with an explicit error since it is not supported by the underlying chat completion backend.
LanceDB python-v0.27.0 adds Voyage multimodal-3.5 embeddings, remote IVF-RQ index support, and parallelized embedding computation.
└──▷ GET THIS VERSION
$ git clone --branch python-v0.27.0 https://github.com/lancedb/lancedb.git
# already have the repo? check out this version:$ git checkout python-v0.27.0
›Exposes table URI via a new table.uri property, giving callers direct access to the underlying storage path.
›Adds support for the voyage-multimodal-3.5 embedding model.
›Supports remote IVF-RQ (Inverted File with Residual Quantization) indexes for vector search on remote tables.
›Parallelizes embedding computations to reduce latency when embedding large batches.
›Enables the HuggingFace embedding feature by default, removing the need to opt in manually.
└──▷ BREAKING ON UPGRADE
!The Rust crate removes default Cargo features (remove default features); any downstream Rust code relying on those features must now enable them explicitly.
LanceDB v0.24.0 adds voyage-multimodal-3.5 embeddings, remote IVF-RQ index support, parallelized embedding computation, and exposes table URI.
└──▷ GET THIS VERSION
$ git clone --branch v0.24.0 https://github.com/lancedb/lancedb.git
# already have the repo? check out this version:$ git checkout v0.24.0
└──▷ USE IT
Inspect the storage URI of an existing table — useful when you need to pass the raw path to another tool or audit where data lives.
python
uri = table.uri
print(uri) # e.g. s3://my-bucket/my-db/my-table.lance
›Exposes table URI via the new table.uri property, letting callers inspect the storage location of a table directly.
›Adds support for voyage-multimodal-3.5 as an embedding model option.
›Supports remote IVF-RQ (Inverted File with Residual Quantization) indexing for remote tables.
›Parallelizes embedding computations to accelerate batch ingestion workflows.
›Enables the huggingface feature flag by default in the Rust crate, removing the need to opt in manually.
└──▷ BREAKING ON UPGRADE
!The Rust crate removes its default features; any crate that relied on default features being enabled must now explicitly list them in its Cargo.toml dependency declaration.
THE AI TOOLCHAIN