The AI Toolchain — issue -195, February 3, 2026

Agno v2.4.8 adds CEL expression support for serializable workflow steps, a Neosantara LLM provider, and a visual Studio editor for AgentOS.

$ git clone --branch v2.4.8 https://github.com/agno-agi/agno.git
# already have the repo? check out this version:
$ git checkout v2.4.8

from agno.agent import Agent
from agno.models.neosantara import Neosantara

agent = Agent(model=Neosantara(id="<model-id>"))
agent.print_response("Halo, apa kabar?")

• ›Adds CEL (Common Expression Language) expression support as evaluators in Condition, Loop, and Router workflow steps, making steps fully serializable as strings.
• ›Adds step_choices support in the Router step's selector function, enabling the router to return a group of steps as a single choice.
• ›Adds Neosantara as a new model provider — an Indonesian LLM gateway with an OpenAI-compatible API.
• ›Adds shebang parsing and Windows command building to Skills for cross-platform script execution.
• ›Introduces Studio: a visual drag-and-drop editor in AgentOS for building Agents, Teams, and Workflows, with a Registry for managing tools, models, databases, and schemas.

PydanticAI v1.52.0 adds OpenAI data-retention control, retry counts in run context, and reasoning-content passthrough.

$ git clone --branch v1.52.0 https://github.com/pydantic/pydantic-ai.git
# already have the repo? check out this version:
$ git checkout v1.52.0

from pydantic_ai.models.openai import OpenAIChatModel

model = OpenAIChatModel('gpt-4o', openai_store=False)

• ›Adds openai_store setting to OpenAIChatModel to control whether OpenAI retains request/response data.
• ›Exposes the number of output-validation retries in the agent's run context, making retry count available to tool and result handlers.
• ›Makes OpenAIChatModel return reasoning content via the same field it was received in, preserving round-trip fidelity of reasoning tokens.

browser-use 0.11.8 adds schema enforcement to the extract tool for structured data extraction.

$ git clone --branch 0.11.8 https://github.com/browser-use/browser-use.git
# already have the repo? check out this version:
$ git checkout 0.11.8

• ›Adds schema enforcement to the extract tool, enabling structured data extraction with a defined extraction schema.
• ›Adds new capabilities to the CLI.

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 v2.1.30 adds PDF page-range reading, OAuth credentials for MCP servers, and a new /debug command

$ git clone --branch v2.1.30 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:
$ git checkout v2.1.30

$ claude mcp add slack-mcp --client-id <your-client-id> --client-secret <your-client-secret>

• ›Adds pages parameter to the Read tool for PDFs, enabling specific page ranges (e.g., pages: "1-5"); large PDFs (>10 pages) now return a lightweight reference when @-mentioned instead of being inlined into context
• ›Adds pre-configured OAuth client credentials for MCP servers that lack Dynamic Client Registration support (e.g., Slack) via new --client-id and --client-secret flags on claude mcp add
• ›Adds /debug command for Claude to help troubleshoot the current session
• ›Supports additional git log and git show flags in read-only mode, including --topo-order, --cherry-pick, --format, and --raw
• ›Adds token count, tool uses, and duration metrics to Task tool results

Continue 1.42.0 adds Nous Research Hermes models and OpenRouter provider support with dynamic model loading.

$ 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]

• ›Adds OpenRouter as a provider with dynamic model loading.
• ›Adds Nous Research Hermes models as a supported provider.

$ 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 v0.39.0 adds prompt history navigation, session rename/delete, collapsible chat blocks, and a transparent UI mode.

$ git clone --branch v0.39.0 https://github.com/charmbracelet/crush.git
# already have the repo? check out this version:
$ git checkout v0.39.0

• ›Adds Up/Down arrow key history navigation to cycle through past prompts, similar to Bash and VSCode.
• ›Enables renaming sessions with Ctrl+R and deleting sessions with Ctrl+X from the session dialog.
• ›Supports click-to-expand/collapse for code blocks, diffs, and thinking blocks in the chat.
• ›Shows last-updated time for each session in the sessions dialog.
• ›Displays the active provider name in the sidebar, panel, and message footers.

GitHub Copilot CLI v0.0.402 adds ACP agent/plan session modes, MCP-to-ACP config bridging, and plugin lifecycle hooks.

$ git clone --branch v0.0.402 https://github.com/github/copilot-cli.git
# already have the repo? check out this version:
$ git checkout v0.0.402

• ›ACP server now supports agent and plan session modes.
• ›MCP configuration now applies to ACP mode, bridging existing MCP setups into ACP workflows.
• ›Plugins can now provide hooks for session lifecycle events.
• ›Plugin update command now works for direct plugins and handles Windows file locks.
• ›Custom agents with unknown fields now load with warnings instead of errors, improving forward compatibility.

$ git clone --branch v0.0.401 https://github.com/github/copilot-cli.git
# already have the repo? check out this version:
$ git checkout v0.0.401

$ copilot login

The open source coding agent.

OpenCode v1.1.49 adds mDNS hostname customization, Haskell formatting, Bedrock Claude Opus caching, SAP AI Core reasoning, and richer desktop/TUI features.

$ git clone --branch v1.1.49 https://github.com/sst/opencode.git
# already have the repo? check out this version:
$ git checkout v1.1.49

$ opencode --mdns-domain mydevbox.local

• ›Adds --mdns-domain flag to customize the mDNS hostname for the OpenCode server.
• ›Adds Ormolu code formatter support for Haskell projects.
• ›Adds prompt caching support for Claude Opus on AWS Bedrock.
• ›Adds reasoning variants support for SAP AI Core provider.
• ›Adds User-Agent header for GitLab AI Gateway requests.

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

Share URLs now default to the pi.dev domain.

$ git clone --branch v0.51.4 https://github.com/earendil-works/pi.git
# already have the repo? check out this version:
$ git checkout v0.51.4

• ›Share URLs now default to pi.dev (replacing the previous default domain).

$ git clone --branch v0.51.3 https://github.com/earendil-works/pi.git
# already have the repo? check out this version:
$ git checkout v0.51.3

$ pi install ./my-extension

$ git clone --branch v0.51.2 https://github.com/earendil-works/pi.git
# already have the repo? check out this version:
$ git checkout v0.51.2

Qwen Code v0.9.0 adds multi-modal input, concurrent batch execution, LSP support, extra_body for OpenAI providers, and a Zed extension.

$ git clone --branch v0.9.0 https://github.com/QwenLM/qwen-code.git
# already have the repo? check out this version:
$ git checkout v0.9.0

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

{
  "contextWindowSize": 32768
}

• ›Adds extra_body support for OpenAI-compatible providers, enabling pass-through of custom request body fields.
• ›Adds contextWindowSize configuration key to control the model context window size.
• ›Adds parentToolCallId and subagentType fields for ACP subagent tracking.
• ›New concurrent runner for batch CLI execution, enabling parallel processing of multiple CLI tasks.
• ›Adds multi-modal input support (image, PDF, audio) across all content generators.

$ git clone --branch v0.9.0-preview.2 https://github.com/QwenLM/qwen-code.git
# already have the repo? check out this version:
$ git checkout v0.9.0-preview.2

contextWindowSize: 32768

$ git clone --branch v0.9.0-nightly.20260203.bd900d36 https://github.com/QwenLM/qwen-code.git
# already have the repo? check out this version:
$ git checkout v0.9.0-nightly.20260203.bd900d36

contextWindowSize: 32768

Get up and running with Kimi-K2.6, GLM-5.2, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.

Ollama v0.15.5 adds sub-agent support, browser-based sign-in, VRAM-aware context limits, and two new models.

$ git clone --branch v0.15.5 https://github.com/ollama/ollama.git
# already have the repo? check out this version:
$ git checkout v0.15.5

$ ollama launch claude -- --resume

• ›Adds argument passthrough to ollama launch, enabling flags like --resume to be forwarded to the launched tool.
• ›Adds sub-agent support to ollama launch for planning, deep research, and similar multi-step agentic workflows.
• ›ollama signin now opens a browser window to the connect page for easier authentication.
• ›Adds GLM-4.7-Flash support on Ollama's experimental MLX engine.
• ›Adds new model Qwen3-Coder-Next, optimized for agentic coding workflows and local development.

Triton v2.65.0 adds first-response latency histogram metrics for coupled models and makes Perf Analyzer dependencies optional.

$ git clone --branch v2.65.0 https://github.com/triton-inference-server/server.git
# already have the repo? check out this version:
$ git checkout v2.65.0

• ›Adds support for the nv_inference_first_response_histogram_ms metric when running models in coupled (non-decoupled) mode, enabling latency histogram tracking for standard inference workloads.
• ›Makes Perf Analyzer dependencies optional when installing tritonclient via pip, reducing install footprint for deployments that don't need performance analysis.
• ›Exposes HTTP errors for non-JSON format responses, improving error visibility for clients sending malformed requests.

llama.cpp b7919 adds Falcon-H1-Tiny-Coder FIM token support to the vocabulary.

$ git clone --branch b7919 https://github.com/ggml-org/llama.cpp.git
# already have the repo? check out this version:
$ git checkout b7919

• ›Adds Falcon-H1-Tiny-Coder Fill-in-the-Middle (FIM) tokens to the vocabulary, enabling FIM-mode code completion with that model.

Phoenix Evals 2.9.0 adds FaithfulnessEvaluator, tool invocation accuracy metric, tool_selection evaluator, and neutral optimization direction.

$ git clone --branch arize-phoenix-evals-v2.9.0 https://github.com/Arize-ai/phoenix.git
# already have the repo? check out this version:
$ git checkout arize-phoenix-evals-v2.9.0

from phoenix.evals import FaithfulnessEvaluator

evaluator = FaithfulnessEvaluator(model=model)
results = evaluator.evaluate(input=query, response=answer, context=retrieved_docs)

• ›Adds FaithfulnessEvaluator class to replace HallucinationEvaluator for grounded response evaluation.
• ›Adds tool_selection evaluator to assess whether an agent selected the correct tool.
• ›Adds tool invocation accuracy metric for measuring correctness of tool calls made by agents.
• ›Adds neutral optimization direction option for Scores, enabling evaluation metrics where neither higher nor lower values are inherently better.
• ›Returns trace_id in Scores objects, enabling direct correlation of evaluation results to traces.

Weave v0.52.26 adds usage stats APIs, a performance flag, Anthropic parse patching, and column ordering for saved views.

$ git clone --branch v0.52.26 https://github.com/wandb/weave.git
# already have the repo? check out this version:
$ git checkout v0.52.26

• ›Adds GET /trace/usage endpoint to dynamically aggregate usage data across traces.
• ›Adds GET /calls/usage endpoint to dynamically bulk-aggregate usage data across calls.
• ›Adds column_order field to SavedViewDefinition to control column ordering in saved views.
• ›Adds patching support for Anthropic parse calls, extending Weave's Anthropic integration.
• ›Introduces a new opt-in flag for improved query performance on distributed ClickHouse clusters (PREWHERE optimization).

LanceDB v0.25.0-beta.0 adds configurable permutation builder memory limits and VoyageAI v4 embedding models.

$ git clone --branch v0.25.0-beta.0 https://github.com/lancedb/lancedb.git
# already have the repo? check out this version:
$ git checkout v0.25.0-beta.0

• ›Allows the permutation builder memory limit to be configured via environment variable.
• ›Adds VoyageAI v4 models as embedding options in the Python client.

$ git clone --branch python-v0.28.0-beta.0 https://github.com/lancedb/lancedb.git
# already have the repo? check out this version:
$ git checkout python-v0.28.0-beta.0