The AI Toolchain — issue -264, November 26, 2025

AutoGPT Platform v0.6.38 adds a Linear Issues block, agent output viewer, GPT-5.1, Claude Opus 4.5, Grok, and Gemini model support.

$ git clone --branch autogpt-platform-beta-v0.6.38 https://github.com/Significant-Gravitas/AutoGPT.git
# already have the repo? check out this version:
$ git checkout autogpt-platform-beta-v0.6.38

• ›Adds a Get Linear Issues Block for retrieving issues from Linear within agent workflows.
• ›Adds an agent outputs viewer in the new builder for inspecting agent run results, with UI/UX improvements.
• ›Adds support for GPT-5.1 and GPT-5.1-codex models.
• ›Adds support for Claude Opus 4.5 model.
• ›Adds support for the latest Grok models.

Your own personal AI assistant. Any OS.

OpenClaw v1.1.0 adds voice-note transcription, media recompression, session intro prompts, and structured command auto-reply payloads.

$ git clone --branch v1.1.0 https://github.com/openclaw/openclaw.git
# already have the repo? check out this version:
$ git checkout v1.1.0

• ›Adds voice-note transcription that feeds a Transcript block and media path into prompts before replying.
• ›Web auto-replies now resize and recompress media, detect media kind, enforce provider caps, and honor the inbound.reply.mediaMaxMb limit.
• ›Supports optional sessionIntro to send the system prompt only once per session, reducing token overhead.
• ›Adds configurable interval for typing-indicator refresh during sessions.
• ›Command auto-replies now return structured {payload, meta} responses, include the command cwd in timeout messages, respect mediaMaxMb for local media, and log Claude metadata.

PydanticAI v1.23.0 adds Anthropic WebFetchTool, cache-message settings, HITL user prompts, and Gemini 3 Pro via OpenRouter.

$ git clone --branch v1.23.0 https://github.com/pydantic/pydantic-ai.git
# already have the repo? check out this version:
$ git checkout v1.23.0

• ›Adds anthropic_cache_messages model setting for Anthropic models, with automatic stripping of cache points that exceed the provider limit.
• ›Adds support for Anthropic's built-in WebFetchTool, enabling web-fetch capability natively through the Anthropic provider.
• ›Allows user_prompt to be supplied in Human-in-the-Loop (HITL) interactions.
• ›Adds Gemini 3 Pro support to OpenRouterModel.
• ›Ensures the openrouter_reasoning model setting is correctly forwarded to the OpenRouter API.

Semantic Kernel Python migrates Google integration to the new Google GenAI SDK.

$ git clone --branch python-1.39.0 https://github.com/microsoft/semantic-kernel.git
# already have the repo? check out this version:
$ git checkout python-1.39.0

• ›Migrates the Google integration to the new Google GenAI SDK, replacing the previous SDK backend.

camel-ai v0.2.80 adds Gemini 3, Claude Opus 4.5, Cerebras, and ERNIE 5.0 Thinking model support alongside workforce running-mode expansion

$ git clone --branch v0.2.80 https://github.com/camel-ai/camel.git
# already have the repo? check out this version:
$ git checkout v0.2.80

• ›Adds support for Gemini 3 models, including thought signatures and streaming mode.
• ›Adds support for Claude Opus 4.5.
• ›Adds Cerebras platform models as a new provider integration.
• ›Adds ERNIE_5_0_THINKING model to the ERNIE model family.
• ›Extends BaseMessage to carry reasoning/thought fields and updates usage accounting in streaming mode.

Continue v1.7.0 removes TTY requirement for headless CLI mode and adds Gemini thought signature support.

$ git clone --branch @continuedev/[email protected] https://github.com/continuedev/continue.git
# already have the repo? check out this version:
$ git checkout @continuedev/[email protected]

• ›Removes TTY requirement for the CLI in headless mode, enabling use in non-interactive environments such as CI pipelines.
• ›Adds support for Gemini thought signature in LLM responses.
• ›Adds support for Claude Opus 4-5 model updates.
• ›Allows renaming a rule before creation in the rule editor.

$ git clone --branch @continuedev/[email protected] https://github.com/continuedev/continue.git
# already have the repo? check out this version:
$ git checkout @continuedev/[email protected]

Glamourous agentic coding for all

Crush v0.19.0 adds Claude Code Max subscription support for Anthropic model access.

$ git clone --branch v0.19.0 https://github.com/charmbracelet/crush.git
# already have the repo? check out this version:
$ git checkout v0.19.0

• ›Supports Claude Code Max subscriptions — select an Anthropic model and Crush will prompt you through the authentication flow.

The open source coding agent.

OpenCode v1.0.115 adds model favorites, PR review comment reactions, diff style control, and git branch display in TUI.

$ git clone --branch v1.0.115 https://github.com/sst/opencode.git
# already have the repo? check out this version:
$ git checkout v1.0.115

diff_style: unified

• ›Adds favorites to the model selector for quick access to preferred models.
• ›Enables reacting to PR Review Comments in GitHub workflows.
• ›Adds diff_style option to allow disabling columnar diffs.
• ›Shows the current git branch in the TUI.
• ›Enables Exa code/websearch integration.

$ git clone --branch v1.0.113 https://github.com/sst/opencode.git
# already have the repo? check out this version:
$ git checkout v1.0.113

Qwen Code v0.3.0-preview.1 adds stream-JSON I/O for headless use and internationalization for UI and LLM output.

$ git clone --branch v0.3.0-preview.1 https://github.com/QwenLM/qwen-code.git
# already have the repo? check out this version:
$ git checkout v0.3.0-preview.1

$ qwen --input-format stream-json --output-format stream-json

• ›Adds stream-json as a supported value for --input-format and --output-format flags, enabling programmatic headless use of Qwen Code via structured streaming JSON.
• ›Adds internationalization (i18n) support for both UI text and LLM output.

$ git clone --branch v0.3.0-nightly.20251126.c9af7481 https://github.com/QwenLM/qwen-code.git
# already have the repo? check out this version:
$ git checkout v0.3.0-nightly.20251126.c9af7481

$ qwen-code --input-format stream-json --output-format stream-json

Zed v0.214.5 adds SSH-remoted agent server extensions, HTTP MCP servers, Vim range writes, and new project_panel.sort_mode and terminal.scrollbar.scroll_multiplier settings.

$ git clone --branch v0.214.5 https://github.com/zed-industries/zed.git
# already have the repo? check out this version:
$ git checkout v0.214.5

{
  "context_servers": {
    "my-remote-server": {
      "enabled": true,
      "url": "http://localhost:3000/mcp"
    }
  }
}

$ :'<,'>w snippet.py

• ›Adds context_servers config support in settings.json for MCP servers connecting over HTTP, with url and enabled fields per server entry.
• ›Adds project_panel.sort_mode setting to control file explorer sort order (directories first, mixed, or files first).
• ›Adds terminal.scrollbar.scroll_multiplier option to tune terminal scrollbar scroll speed.
• ›Adds diffs_expanded key context to the Editor key context when diff changes are visible, enabling keybinding rules that target diff state.
• ›Makes Agent Server Extensions available over SSH remoting, including per-target environment variable support.

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

gemini-cli v0.19.0-nightly adds /resume slash command, model availability service setting, and Flash quota fallback.

$ git clone --branch v0.19.0-nightly.20251126.03845198c https://github.com/google-gemini/gemini-cli.git
# already have the repo? check out this version:
$ git checkout v0.19.0-nightly.20251126.03845198c

$ /resume

• ›New /resume slash command opens an interactive session browser to restore previous chats.
• ›New isModelAvailabilityServiceEnabled config setting controls the model availability service.
• ›Automatically falls back to Gemini Flash when a TerminalQuota error is encountered.

$ git clone --branch v0.18.0 https://github.com/google-gemini/gemini-cli.git
# already have the repo? check out this version:
$ git checkout v0.18.0

$ git clone --branch v0.19.0-preview.0 https://github.com/google-gemini/gemini-cli.git
# already have the repo? check out this version:
$ git checkout v0.19.0-preview.0

$ /resume

$ /stats

KoboldCpp v1.102.3 adds llama.cpp UI, jinja2 template modes, Qwen3Next support, and universal tool-calling improvements.

$ git clone --branch v1.102.3 https://github.com/LostRuins/koboldcpp.git
# already have the repo? check out this version:
$ git checkout v1.102.3

$ koboldcpp --model mymodel.gguf --jinja

$ koboldcpp --testmemory

• ›Adds --jinja flag to use jinja2 templates from GGUF in /v1/chat/completions for normal messages while retaining the KoboldCpp universal tool-calling module.
• ›Adds --jinja_tools flag to use jinja2 templates from GGUF for all messages and tool calls in /v1/chat/completions, giving the model and frontend full responsibility for compatibility.
• ›Adds --testmemory debug flag to inspect values retrieved by automatic GPU detection.
• ›Adds sampler support for smoothing curve.
• ›Bundles the llama.cpp UI as an extra interface accessible at http://localhost:5001/lcpp, supporting document imports (.pdf) alongside the existing KoboldAI Lite UI.

LocalAI v3.8.0 adds universal model import from HF/Ollama/OCI, MCP live streaming, hot-reloadable settings, logprobs/logitbias, and new llama.cpp YAML options.

$ git clone --branch v3.8.0 https://github.com/mudler/LocalAI.git
# already have the repo? check out this version:
$ git checkout v3.8.0

options:
- context_shift:false
- cache_ram:-1
- use_jinja:true
- parallel:2
- grpc_servers:localhost:50051,localhost:50052

• ›Adds context_shift, cache_ram, use_jinja, parallel, and grpc_servers as tunable options under the options: key in the model YAML config for llama.cpp backends.
• ›Adds a new LocalAI endpoint to stream live MCP agent results — tool calls, reasoning steps, and intermediate actions are now visible in real-time via the Model Context Protocol.
• ›Adds full support for logprobs and logitbias per the OpenAI specification, enabling token-level probability data for agentic logic, Self-RAG, and hallucination evaluation.
• ›Introduces universal model import from Hugging Face, Ollama, OCI registries, or local file paths via URL — auto-detects backends (e.g., llama.cpp vs diffusers) and applies native chat templates (e.g., llama-3, mistral) from model metadata.
• ›Adds a hot-reloadable 'Settings > System' panel for toggling watchdogs, P2P, gallery availability, and API key management (generate, rotate, expire) without restarting — requires mounting /configuration in containerized deployments.

Triton v2.63.0 adds v1/embeddings endpoint for vLLM and echo parameter support in the OpenAI-compatible API server.

$ git clone --branch v2.63.0 https://github.com/triton-inference-server/server.git
# already have the repo? check out this version:
$ git checkout v2.63.0

• ›Enables the v1/embeddings endpoint for the vLLM backend in the OpenAI-compatible API server.
• ›Enables the echo parameter for TensorRT-LLM and Python backends in the OpenAI-compatible API server.
• ›Provides more specific, OpenAI-compliant error codes from the OpenAI-compatible API server.

Weave v0.52.20 adds Bedrock Agents tracing, TypeScript withAttributes, OpenAI x-request-id tracking, prompt vars in streaming completions, and ObjectRef.fromUri.

$ git clone --branch v0.52.20 https://github.com/wandb/weave.git
# already have the repo? check out this version:
$ git checkout v0.52.20

import { withAttributes } from '@wandb/weave';

await withAttributes({ environment: 'prod', version: '2.1.0' }, async () => {
  // traced code here
});

• ›Adds completions_create_stream support for passing prompt and template vars directly, enabling prompt-managed streaming completions.
• ›Implements ObjectRef.fromUri method to construct ObjectRef instances from Weave URIs.
• ›Tracks x-request-id headers from the OpenAI Responses API, surfacing request correlation IDs in traces.
• ›Adds withAttributes to the TypeScript SDK for attaching arbitrary attributes to spans.
• ›Adds integration tracing for AWS Bedrock Agents.