Heads up This site is currently under heavy development.
Subscribe Get it delivered — the daily firehose, filtered to the tools you run, plus the documentation changes vendors never announce. Compare plans →

The AI Toolchain — issue -194, February 4, 2026

THE AI TOOLCHAIN NO. -194
Tail
THE DAILY RELEASE FIREHOSE
PUBLISHED FEBRUARY 4, 2026 · EVERY WEEKDAY
EDITIONS tail grep head diff uniq

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.

VIEW
ISSUE VIEW full issue
Do you prefer this view?
$ tct list   # 13 tools matched
AI & LLM Tooling
◆  AI Agent Frameworks

AutoGPT

Sources Release notes → autogpt-platform-beta-v0.6.46 NOTES

AutoGPT Platform adds persistent CoPilot workspace, Whisper speech-to-text, Redis SSE reconnection, and a marketplace agent customization tool.

└──▷ GET THIS VERSION
$ git clone --branch autogpt-platform-beta-v0.6.46 https://github.com/Significant-Gravitas/AutoGPT.git
# already have the repo? check out this version:
$ git checkout autogpt-platform-beta-v0.6.46
  • Adds customize_agent tool enabling customization of marketplace agent templates directly within the platform.
  • Adds Redis-based SSE reconnection to keep long-running CoPilot operations alive across dropped connections.
  • Adds User Workspace for persistent file storage across CoPilot sessions.
  • Adds Whisper-model speech-to-text input for CoPilot conversations.
  • Adds two-phase library agent fetching to support sub-agent discovery and selection.
+1 moreshow less
  • Covers webhook ingress URL routing for inbound webhook handling.
└──▷ BREAKING ON UPGRADE
  • !The Claude 3.7 Sonnet model is removed; existing agents using it are migrated via the included migration and defensive handling.
Was this useful?

OpenClaw

Sources Release notes → v2026.2.2 NOTES

Your own personal AI assistant. Any OS.

OpenClaw v2026.2.2 adds Feishu/Lark integration, an Agents dashboard, QMD memory backend, and configurable subagent thinking levels.

└──▷ GET THIS VERSION
$ git clone --branch v2026.2.2 https://github.com/openclaw/openclaw.git
# already have the repo? check out this version:
$ git checkout v2026.2.2
└──▷ USE IT
Set a default thinking level for all subagents, overriding per-agent as needed — useful for tuning reasoning depth across a fleet.
yaml
# In your OpenClaw config file:
agents:
  defaults:
    subagents:
      thinking: high
  list:
    - name: fast-agent
      subagents:
        thinking: low
  • Adds Feishu/Lark plugin support, enabling OpenClaw agents to operate in Feishu workspaces.
  • New Agents dashboard in the Web UI for managing agent files, tools, skills, models, channels, and cron jobs.
  • Introduces opt-in QMD backend for workspace memory storage.
  • Adds a healthcheck skill and bootstrap audit guidance for security posture validation.
  • Supports setting a default subagent thinking level via agents.defaults.subagents.thinking (and per-agent agents.list[].subagents.thinking).
Was this useful?

PydanticAI

Sources Release notes → v1.54.0 2 RELEASES · 2026-02-04 NOTES STABLE

PydanticAI v1.54.0 adds concurrency limiting for Agents and Models

└──▷ GET THIS VERSION
$ git clone --branch v1.54.0 https://github.com/pydantic/pydantic-ai.git
# already have the repo? check out this version:
$ git checkout v1.54.0
  • Adds concurrency limiting for Agents and Models to cap parallel executions and prevent resource exhaustion.
1 more release in this issue · 2026-02-04
v1.53.0 NOTES STABLE

PydanticAI v1.53.0 automatically infers the gateway base URL from the token region.

└──▷ GET THIS VERSION
$ git clone --branch v1.53.0 https://github.com/pydantic/pydantic-ai.git
# already have the repo? check out this version:
$ git checkout v1.53.0
  • Automatically infers the gateway base URL from the token region, removing the need to manually specify it.
Was this useful?
◆  AI Coding Agents

Anthropic Claude Code

Sources Release notes → v2.1.31 NOTES

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 v2.1.31 adds session resume hints and Japanese IME support in checkbox selection.

└──▷ GET THIS VERSION
$ git clone --branch v2.1.31 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:
$ git checkout v2.1.31
  • Adds a session resume hint on exit, showing the command to continue a conversation later.
  • Supports full-width (zenkaku) space input from Japanese IME in checkbox selection.
Was this useful?

Continue

Sources Release notes → v1.2.15-vscode NOTES

Continue v1.2.15 adds cn review, subagents, agent skills, OpenRouter, Ask Sage, and tool prompt overrides via .continuerc.json.

└──▷ GET THIS VERSION
$ git clone --branch v1.2.15-vscode https://github.com/continuedev/continue.git
# already have the repo? check out this version:
$ git checkout v1.2.15-vscode
└──▷ USE IT
Override the prompt for a specific tool in a project without editing the global config.
json
{
  "tools": {
    "run_terminal_command": {
      "prompt": "Only run read-only commands. Never modify files."
    }
  }
}
  • Adds cn review CLI subcommand (previously cn check) for AI agent code review with UX improvements.
  • Adds tool prompt override support in .continuerc.json, letting users customize tool prompts at the project level.
  • Allows MCP/Bash tools by default in headless mode, removing manual opt-in for agent automation workflows.
  • Adds subagents support, enabling agents to spawn and coordinate child agents.
  • Adds agent skills feature (in both VS Code extension and CLI) for composable, reusable agent capabilities.
+8 moreshow less
  • Adds OpenRouter provider with dynamic model loading.
  • Adds Ask Sage as a supported provider for the Continue CLI.
  • Adds new OVHcloud models as a provider.
  • Adds Nous Research Hermes models as a provider.
  • Adds a remove rule button in the UI for managing active rules.
  • Adds automatic resubmission on overloaded/rate-limit errors from providers.
  • Adds message counter display on the history page.
  • Adds WSL detection in the CLI to spawn the appropriate shell automatically.
Was this useful?

Block Goose

Sources Release notes → v1.23.0 NOTES

an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM

Goose v1.23.0 adds WebSocket and Streamable HTTP transports for ACP, Laminar observability integration, and a CLI flag to skip profile extensions.

└──▷ GET THIS VERSION
$ git clone --branch v1.23.0 https://github.com/block/goose.git
# already have the repo? check out this version:
$ git checkout v1.23.0
  • Adds WebSocket transport support for goose-acp agent communication.
  • Adds Streamable HTTP transport for ACP and goose-acp usage.
  • Integrates Laminar for agent observability and tracing.
  • New CLI flag to skip loading profile extensions at startup.
  • Adds a thinking status hook for surfacing model reasoning state.
+3 moreshow less
  • Adds hourly skill fetching to keep skills automatically up to date.
  • Introduces a platform extension migrator alongside a 'code mode' rename.
  • Adds an Ask AI Discord bot integration.
Was this useful?

OpenAI Codex CLI

Sources Release notes → rust-v0.96.0 2 RELEASES · 2026-02-04 NOTES STABLE

Lightweight coding agent that runs in your terminal

Codex CLI v0.96.0 adds thread compaction API, WebSocket rate-limit signaling, and unified exec on all non-Windows platforms.

└──▷ GET THIS VERSION
$ git clone --branch rust-v0.96.0 https://github.com/openai/codex.git
# already have the repo? check out this version:
$ git checkout rust-v0.96.0
  • Adds thread/compact as an async trigger RPC in the v2 app-server API, letting clients start compaction immediately and track completion separately.
  • Adds a codex.rate_limits WebSocket event for rate-limit signaling, with WebSocket parity for ETag and reasoning metadata handling.
  • Enables unified_exec on all non-Windows platforms.
  • Constrained requirement values now include source provenance, enabling source-aware config debugging in UI flows such as /debug-config.
1 more release in this issue · 2026-02-04
rust-v0.95.0 NOTES STABLE

Codex CLI adds macOS app launcher, personal skills, parallel shell tools, and CODEX_THREAD_ID injection.

└──▷ GET THIS VERSION
$ git clone --branch rust-v0.95.0 https://github.com/openai/codex.git
# already have the repo? check out this version:
$ git checkout rust-v0.95.0
└──▷ TRY IT
Invoke /plan with an inline prompt in the TUI instead of typing it separately after the command.
$ /plan refactor the authentication module to use JWT tokens
Use CODEX_THREAD_ID in a skill or script to branch logic based on which Codex session is active.
$ echo "Running in Codex thread: $CODEX_THREAD_ID"
  • Adds codex app <path> command on macOS to launch Codex Desktop from the CLI, auto-downloading the DMG if not present.
  • Supports personal skill loading from ~/.agents/skills (with ~/.codex/skills compatibility) and new app-server APIs/events to list and download public remote skills.
  • /plan now accepts inline prompt arguments and pasted images directly in the TUI, with improved slash-command editing and highlighting.
  • Shell-related tools can now execute in parallel, improving multi-command throughput.
  • Injects CODEX_THREAD_ID into the shell environment so scripts and skills can detect the active thread/session.
+1 moreshow less
  • Adds vendored Bubblewrap with FFI wiring in the Linux sandbox as groundwork for upcoming runtime integration.
Was this useful?

SST OpenCode

Sources Release notes → v1.1.51 2 RELEASES · 2026-02-04 NOTES STABLE

The open source coding agent.

OpenCode v1.1.51 restores OSC52 clipboard support and prioritizes OPENCODE_CONFIG_DIR for AGENTS.md lookup.

└──▷ GET THIS VERSION
$ git clone --branch v1.1.51 https://github.com/sst/opencode.git
# already have the repo? check out this version:
$ git checkout v1.1.51
  • Prioritizes OPENCODE_CONFIG_DIR environment variable when resolving the AGENTS.md file location.
  • Restores direct OSC52 clipboard support in the TUI.
1 more release in this issue · 2026-02-04
v1.1.50 NOTES STABLE

OpenCode v1.1.50 adds shell.env hook, Trinity model support, skills directory loading, and --thinking flag for reasoning blocks.

└──▷ GET THIS VERSION
$ git clone --branch v1.1.50 https://github.com/sst/opencode.git
# already have the repo? check out this version:
$ git checkout v1.1.50
└──▷ TRY IT
See reasoning blocks inline when running a non-interactive prompt — useful for auditing how the model arrived at an answer.
$ opencode run --thinking "Explain the authentication flow in this codebase"
Inject or override environment variables for all tool and shell invocations without modifying your shell profile.
javascript
// In your opencode plugin (shell.env hook)
export default {
  hooks: {
    "shell.env": (env) => ({
      ...env,
      MY_API_KEY: "secret",
      NODE_ENV: "production",
    }),
  },
};
  • Adds shell.env hook for manipulating environment variables in tools and shell via plugins.
  • Adds Trinity model system prompt support.
  • Supports reading skills from .agents/skills directories.
  • Adds --thinking flag to the run command to display reasoning blocks.
  • Allows theme colors in agent customization.
+4 moreshow less
  • Adds .slnx to C#/F# LSP root detection.
  • Makes the CLI run command non-interactive.
  • Moves session search to the command palette in the desktop app.
  • Allows empty prompt with review comments in the desktop app.
Was this useful?

Earendil Works Pi

Sources Release notes → v0.51.6 NOTES

AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI

Pi v0.51.6 adds a configurable resume keybinding and smarter slash command triggering.

└──▷ GET THIS VERSION
$ git clone --branch v0.51.6 https://github.com/earendil-works/pi.git
# already have the repo? check out this version:
$ git checkout v0.51.6
└──▷ USE IT
Bind a key to open the session resume selector so you can quickly return to a previous session without reaching for the mouse.
json
# In your keybindings config (see docs/keybindings.md)
{ "key": "ctrl+r", "action": "resume" }
  • Adds resume as a configurable keybinding action to open the session resume selector, on par with newSession, tree, and fork.
  • Slash command menu now triggers on the first line even when other lines contain content, enabling commands to be prepended to existing text.
Was this useful?

Alibaba Qwen Code

Sources Release notes → v0.9.1-preview.0 3 RELEASES · 2026-02-04 NOTES STABLE

Qwen Code v0.9.1-preview.0 adds symlink support for skill manager and preserves UTF-8 BOM on file edits.

└──▷ GET THIS VERSION
$ git clone --branch v0.9.1-preview.0 https://github.com/QwenLM/qwen-code.git
# already have the repo? check out this version:
$ git checkout v0.9.1-preview.0
  • Adds symlink support for the skill manager, allowing skills to be referenced via symbolic links.
  • Preserves UTF-8 BOM when editing files, preventing BOM stripping on save for files that require it.
  • Adds cache_control to system and last user text messages in the Anthropic content generator, reducing prompt-processing costs on repeated context.
2 more releases in this issue · 2026-02-04
v0.9.0-preview.3 NOTES STABLE

Qwen Code v0.9.0-preview.3 adds symlink support for the skill manager.

└──▷ GET THIS VERSION
$ git clone --branch v0.9.0-preview.3 https://github.com/QwenLM/qwen-code.git
# already have the repo? check out this version:
$ git checkout v0.9.0-preview.3
  • Adds symlink support for the skill manager, enabling skills to be referenced via symbolic links.
v0.9.0-nightly.20260204.7fb45c56 NOTES STABLE

Qwen Code v0.9.0-nightly adds symlink support for the skill manager.

└──▷ GET THIS VERSION
$ git clone --branch v0.9.0-nightly.20260204.7fb45c56 https://github.com/QwenLM/qwen-code.git
# already have the repo? check out this version:
$ git checkout v0.9.0-nightly.20260204.7fb45c56
  • Adds symlink support for the skill manager, allowing skill files to be referenced via symbolic links.
Was this useful?

Zed

Sources Release notes → v0.222.2 NOTES

Zed v0.222.2 adds multi-provider edit predictions, Windows SSH remoting, image viewer zoom/pan, and new vim/config settings.

└──▷ GET THIS VERSION
$ git clone --branch v0.222.2 https://github.com/zed-industries/zed.git
# already have the repo? check out this version:
$ git checkout v0.222.2
└──▷ USE IT
Pin vim.gdefault so all substitute commands replace every match on a line without needing /g each time.
json
{
  "vim.gdefault": true
}
Auto-dismiss noisy LSP server notifications after 5 seconds instead of leaving them on screen indefinitely.
json
{
  "global_lsp_settings": {
    "notifications": {
      "dismiss_timeout_ms": 5000
    }
  }
}
Install a specific Zed version in CI or on a new machine instead of always pulling latest.
$ ZED_VERSION=0.222.2 sh <(curl -fsSL https://zed.dev/install.sh)
  • Adds vim.gdefault setting (plus :set gdefault, :set nogdefault, short forms gd/nogd) to make /g replace-all the default for substitutions.
  • Adds project_panel.bold_folder_labels setting (default false) to render folder names in bold in the project panel.
  • Adds global_lsp_settings.notifications.dismiss_timeout_ms setting to auto-dismiss server notifications after a configurable timeout (e.g. 5000 ms).
  • Adds ZED_VERSION environment variable to the install script to pin a specific Zed version (defaults to latest).
  • Adds support for multiple edit prediction providers — GitHub Copilot Next Edit Suggestions, Ollama, Codestral, Sweep, and Mercury Coder — configurable via the Edit Prediction menu.
+19 moreshow less
  • Adds Windows as a supported target platform for SSH remoting.
  • Adds zooming and panning to the image viewer with zoom in/out, reset, fit-to-view, zoom-to-actual actions, scroll-wheel zoom with modifier, click-and-drag panning, zoom percentage overlay, and toolbar controls.
  • Adds editor: select to start of larger syntax node and editor: select to end of larger syntax node commands for extending selections to syntax node boundaries.
  • Adds ability to rename terminal tabs via right-click context menu and double-click.
  • Adds configurable remote server cache retention (default 5).
  • Adds support for importing VS Code Insiders and VSCodium Insiders user settings (settings.json).
  • Adds 'Reopen with Encoding' feature for local files, allowing buffers to be re-read with a different encoding.
  • Adds a menu item to open Markdown and SVG files in the preview tab from the buffer and tab right-click menu.
  • Adds 'Reveal in File Manager' to the project panel context menu on WSL.
  • Adds a menu item to close the current project and return to a clean window state.
  • Adds a setting on the terminal tool call card stop button to control whether it cancels the running command only, or the entire thread.
  • Adds relative line jump support to the go-to-line action via +/-/f/b/F/B in Vim/Helix mode.
  • Adds first-line language detection for extensionless libc++ headers in C++.
  • Adds syntax highlights for #elifdef and #elifndef preprocessor directives and document comments in C/C++.
  • Adds a context menu in the agent panel for inserting context items with the mouse.
  • Makes the message queue feature available for external agents.
  • Adds support for specifying multiple snippet paths in extensions.
  • Significantly reduces resource usage when watching filesystem changes on macOS.
  • Git panel project picker now displays Git status icons to indicate which repos have uncommitted changes.
Was this useful?

Google gemini-cli

Sources Release notes → v0.27.0 NOTES

An open-source AI agent that brings the power of Gemini directly into your terminal.

gemini-cli v0.27.0 adds /rewind, /agents config, MCP enable/disable, Plan Mode, AskUser tool, image paste on Linux, and more.

└──▷ GET THIS VERSION
$ git clone --branch v0.27.0 https://github.com/google-gemini/gemini-cli.git
# already have the repo? check out this version:
$ git checkout v0.27.0
└──▷ TRY IT
Roll back a conversation to an earlier turn when the agent went down the wrong path.
$ /rewind
Enable or disable an MCP server at runtime without editing config files.
$ /mcp enable <server-name>
Inspect and configure subagents discovered by the AgentRegistry.
$ /agents config
  • Adds /rewind command to roll back conversation history to a previous state.
  • Adds /agents config command for configuring and discovering subagents.
  • Adds enable/disable commands for MCP servers via /mcp enable and /mcp disable.
  • Adds AskUser tool and accompanying AskUserDialog UI for structured mid-task user prompts.
  • Adds Plan mode, now reachable via Shift+Tab mode cycling, with its own UI theme and a communicate tool kind.
+15 moreshow less
  • Adds code-reviewer skill for automated code review workflows.
  • Adds docs-writer skill for documentation authoring workflows.
  • Enables image pasting on Linux (Wayland and X11).
  • Adds dynamic variable substitution in system prompt overrides.
  • Adds ANSI escape sequence sanitization in non-interactive (piped/redirect) output.
  • Enables the event-driven tool execution scheduler by default.
  • Enables the hooks system by default, with hook-injected context wrapped in distinct XML tags for security.
  • Adds AgentRegistry to track all discovered subagents across extensions.
  • Replaces large text pastes with a [Pasted Text: X lines] placeholder to keep context concise.
  • Adds JetBrains IDE detection for IDE-companion integration.
  • Displays user tier (e.g., free/paid) in the /about command output.
  • Consolidates shell inactivity and redirection monitoring into a unified monitor.
  • Disables startup tips automatically after 10 runs.
  • Adds an agent configuration dialog with full state management.
  • Adds A/B test routing using numerical complexity scoring for Gemini model selection.
└──▷ BREAKING ON UPGRADE
  • !The default behavior for positional arguments changes to interactive mode (Default to interactive mode for positional arguments), which may break scripts relying on the previous non-interactive default.
  • !Legacy settings removed (feat(core): Remove legacy settings) — configurations relying on those settings will no longer be applied.
Was this useful?
◆  Local LLM Runtimes

vLLM

Sources Release notes → v0.15.1 NOTES

vLLM v0.15.1 adds Step-3.5-Flash model support and restores RTX Blackwell GPU compatibility.

└──▷ GET THIS VERSION
$ git clone --branch v0.15.1 https://github.com/vllm-project/vllm.git
# already have the repo? check out this version:
$ git checkout v0.15.1
  • Adds support for the Step-3.5-Flash model.
Was this useful?
my-toolchain — 0 tools
paste an install list to detect your tools

A brew list, a Brewfile, requirements.txt, a Dockerfile — or just the product names, free-form. Nothing leaves your browser.

    browse all tools →