Backfill issue -176 — The AI Toolchain

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   # 3 tools matched

AI & LLM Tooling

◆ AI Coding Agents

Charm Crush

v0.44.0 NOTES

Glamourous agentic coding for all

Crush v0.44.0 adds visual confirmation when skills are loaded so you know the LLM isn't improvising.

└──▷ GET THIS VERSION

$ git clone --branch v0.44.0 https://github.com/charmbracelet/crush.git
# already have the repo? check out this version:
$ git checkout v0.44.0

›Adds UI indicator showing when a skill is actively loaded, distinguishing real skill execution from LLM improvisation.

Was this useful?

Alibaba Qwen Code

sdk-typescript-v0.1.5 NOTES

Qwen Code SDK TypeScript v0.1.5 adds session resumption, multi-modal input, concurrent batch runner, LSP support, and a Zed extension.

└──▷ GET THIS VERSION

$ git clone --branch sdk-typescript-v0.1.5 https://github.com/QwenLM/qwen-code.git
# already have the repo? check out this version:
$ git checkout sdk-typescript-v0.1.5

└──▷ USE IT

Inject provider-specific parameters (e.g. thinking budget) to an OpenAI-compatible provider via extra_body in settings.

json

{
  "extra_body": {
    "enable_thinking": true,
    "thinking_budget": 8192
  }
}

Set environment variables for tool use directly inside the settings file instead of the shell environment.

json

{
  "env": {
    "MY_API_KEY": "sk-...",
    "CUSTOM_ENDPOINT": "https://my-proxy.example.com"
  }
}

›Adds --session-id CLI flag and resume/continue options in the SDK to resume or continue existing sessions.
›Adds settings.env field in settings configuration for environment variable configuration.
›Adds contextWindowSize configuration support.
›Adds extra_body support for OpenAI-compatible providers.
›Adds parentToolCallId and subagentType fields for ACP subagent tracking.

+21 moreshow less

›Adds FORK_MODE support to ProcessTransport for Electron IPC integration.
›Adds an export command for session history with Markdown and HTML output formats.
›Adds a concurrent runner for batch CLI execution.
›Adds multi-modal input support (image, PDF, audio) across all content generators.
›Adds experimental LSP support for code intelligence.
›Adds a Zed extension for the Qwen Code agent server.
›Adds MCP tool progress update support in TUI and SDK mode.
›Adds Coding Plan authentication mode with a unified AuthDialog and configuration management.
›Adds symlink support for the skill manager.
›Promotes Agent Skills from experimental to stable.
›Adds cache_control for system and last user text messages in the Anthropic content generator to optimize prompt caching.
›Adds TPM throttling error handling with a 1-minute retry delay.
›Adds source information tracking in telemetry logs.
›Improves retry logic for better 429/5xx error handling.
›Refactors debug mode output to route console calls to a logfile-first debugLogger.
›Adds large paste placeholder and improves enter-submit behavior on macOS.
›Supports model selection through ACP in the VS Code IDE companion.
›Adds hint for installing external source extensions.
›Implements unified UI architecture with shared component library in the web UI.
›Clarifies output formats for non-interactive mode.
›Preserves UTF-8 BOM when editing files.

└──▷ BREAKING ON UPGRADE

!The read_many_files tool is removed; its functionality is replaced by a readManyFiles utility available via user @-commands.
!The Smart Edit tool and ClearcutLogger are removed.
!Settings keys previously named with disable* prefix are renamed to enable* positive naming.

Was this useful?

◆ Local LLM Runtimes

llama.cpp

b8128 NOTES

llama.cpp b8128 adds support for the Kanana-2 model architecture.

└──▷ GET THIS VERSION

$ git clone --branch b8128 https://github.com/ggml-org/llama.cpp.git
# already have the repo? check out this version:
$ git checkout b8128

›Adds support for the Kanana-2 model architecture.

Was this useful?

// END OF TRANSMISSION

The AI Toolchain · curated daily · February 22, 2026
Releases & features from the open-source and commercial AI tools that practitioners actually run.

The AI Toolchain — issue -176, February 22, 2026

Charm Crush

Alibaba Qwen Code

llama.cpp