Backfill issue -289 — 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   # 6 tools matched

AI & LLM Tooling

◆ AI Agent Frameworks

Agno (formerly Phidata)

v2.2.6 NOTES

Agno v2.2.6 adds conversational workflows, a Notion toolkit, model-as-string syntax, and session state on run events.

└──▷ GET THIS VERSION

$ git clone --branch v2.2.6 https://github.com/agno-agi/agno.git
# already have the repo? check out this version:
$ git checkout v2.2.6

└──▷ USE IT

Reference a model by string instead of instantiating a model object — useful for quickly switching providers in agent definitions.

python

from agno.agent import Agent

agent = Agent(model="openai:gpt-5")

›Supports defining models as a string (e.g., openai:gpt-5) when configuring Agents and Teams, eliminating the need to instantiate a model object.
›Adds session state access on RunOutput and RunCompleted events for Agents, and on TeamRunOutput and TeamRunCompleted events for Teams.
›New Notion Toolkit lets Agents read and interact with Notion pages.
›New Conversational Workflows capability gives Workflows a chat-like experience similar to Agent and Team, including session and history support.
›Adds input schema validation for Agents and Teams on AgentOS.

+3 moreshow less

›Extends FileTools toolkit with chunked reading of large files, partial updates to large files, and file deletion.
›Adds reranker support to the Milvus vector database search operation, reordering results by relevance after initial vector search.
›All model implementations now cache and reuse HTTP clients (client persistence) to reduce connection overhead.

Was this useful?

CrewAI

1.3.0 NOTES

Framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks.

CrewAI 1.3.0 enhances the QdrantVectorSearchTool with improved vector search capabilities.

└──▷ GET THIS VERSION

$ git clone --branch 1.3.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.3.0

›Enhances QdrantVectorSearchTool with expanded capabilities for vector-based semantic search.

Was this useful?

browser-use

0.9.5 NOTES

browser-use 0.9.5 adds a template system, remote code upload/execution, and extra_body support for OpenRouter

└──▷ GET THIS VERSION

$ git clone --branch 0.9.5 https://github.com/browser-use/browser-use.git
# already have the repo? check out this version:
$ git checkout 0.9.5

›Adds extra_body parameter support for OpenRouter LLM calls, enabling pass-through of provider-specific request fields.
›Implements a template system backed by a template-library repo, letting users bootstrap agent tasks from pre-built templates.
›Adds remote code upload and execution capability, enabling agents to run code on remote targets.
›Improves cloud parameters configuration for browser-use cloud deployments.
›Adds Unicode character support in filenames and improved path handling, including Windows path validation.

Was this useful?

◆ Local LLM Runtimes

KoboldCpp

v1.101.1 NOTES

KoboldCpp v1.101.1 adds Qwen3-VL vision support, Qwen Image/Image-Edit generation, logprobs streaming, VITS API endpoint, and Kobold Lite branching.

└──▷ GET THIS VERSION

$ git clone --branch v1.101.1 https://github.com/LostRuins/koboldcpp.git
# already have the repo? check out this version:
$ git checkout v1.101.1

└──▷ TRY IT

Pass multiple key-value overrides in one launch command to fine-tune model behavior without repeating the flag.

$ koboldcpp --model mymodel.gguf --overridekv rope_freq_base=500000,rope_freq_scale=1.0

›Adds support for multiple --overridekv values split by commas in a single invocation.
›Renames --blasbatchsize to --batchsize (old name remains functional as an alias).
›Adds aliases for OpenAI-compatible endpoints without the /v1/ prefix.
›Adds logprobs streaming for the OpenAI chat completions API (values sent at end of stream).
›Adds a VITS API server compatibility endpoint.

+10 moreshow less

›Adds an image generation scheduler selection option in SDUI.
›Supports Qwen3-VL multimodal vision-language models (load model + mmproj as usual).
›Adds Qwen Image and Qwen Image Edit generation support: load the Qwen Image Edit model as the image gen model, the Qwen Image VAE as VAE, Qwen2.5-VL-7B-Instruct as Clip-1, and its mmproj as Clip-2; image editing available via http://localhost:5001/sdui by uploading a reference image.
›Adds Horde worker job stats.
›Adds branching support in Kobold Lite, allowing multiple story branches (create, delete, and swap between branches) within a single save file.
›Adds ccv3 tavern card support in Kobold Lite.
›Upgrades bundled ROCm library in the ROCm Linux binary to 7.1, enabling AMD Radeon 9000-series GPU support on Linux without compiling from source.
›Adds experimental fractional scaling support in the GUI launcher for Wayland on GNOME.
›Adds right-click copy on the SDUI image preview.
›Makes GPU layer count preview in the GUI more accurate (removes the previous +2 layer offset).

Was this useful?

◆ AI Coding Agents

SST OpenCode

v1.0.9 2 RELEASES · 2025-11-01 NOTES STABLE

The open source coding agent.

OpenCode v1.0.9 adds /exit command and new aliases for TUI commands including /summarize, /clear, and /resume.

└──▷ GET THIS VERSION

$ git clone --branch v1.0.9 https://github.com/sst/opencode.git
# already have the repo? check out this version:
$ git checkout v1.0.9

└──▷ TRY IT

Exit the TUI cleanly from the command prompt instead of using a keyboard interrupt.

$ /exit

Resume a previous session using the more intuitive /continue alias.

$ /continue

›Adds /exit command (with aliases /quit and /q) to quit the TUI without reaching for Ctrl-C.
›Adds /summarize as an alias for /compact, /clear as an alias for /new, and /resume//continue as aliases for /session.
›Extends TUI autocomplete to recognise all new command aliases.

1 more release in this issue · 2025-11-01

v1.0.8 NOTES STABLE

OpenCode v1.0.8 adds --prompt CLI option, sidebar position memory, formatter status display, and share URL toast notifications.

└──▷ GET THIS VERSION

$ git clone --branch v1.0.8 https://github.com/sst/opencode.git
# already have the repo? check out this version:
$ git checkout v1.0.8

└──▷ TRY IT

Kick off a coding session with a pre-filled prompt without typing it interactively.

$ opencode --prompt "Refactor the auth module to use JWT tokens"

›Adds --prompt option to the TUI for passing a prompt directly at launch.
›Remembers sidebar position across sessions in the TUI.
›Displays formatter status in the TUI status dialog.
›Shows a toast notification when a /share URL is copied to clipboard.
›Makes /init a default slash command on the server side.

Was this useful?

Google gemini-cli

v0.13.0-nightly.20251101.caf2ca14 NOTES

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

gemini-cli v0.13.0-nightly adds Kitty function-key support, release channel detection, and untrusted-folder handling on extension install.

└──▷ GET THIS VERSION

$ git clone --branch v0.13.0-nightly.20251101.caf2ca14 https://github.com/google-gemini/gemini-cli.git
# already have the repo? check out this version:
$ git checkout v0.13.0-nightly.20251101.caf2ca14

›Adds Kitty terminal support for function keys, enabling full keyboard navigation in Kitty-based environments.
›Introduces release channel detection so the CLI can identify and report which release channel (stable, nightly, etc.) is in use.
›Handles untrusted folders during extension install and link, prompting appropriately rather than silently proceeding.
›Adds extension name to ToolCallEvent telemetry, giving operators visibility into which extension triggered each tool call.
›Adds slow-rendering metric and logging to telemetry, enabling detection of UI performance regressions.

Was this useful?

// END OF TRANSMISSION

The AI Toolchain · curated daily · November 1, 2025
Releases & features from the open-source and commercial AI tools that practitioners actually run.

The AI Toolchain — issue -289, November 1, 2025

Agno (formerly Phidata)

CrewAI

browser-use

KoboldCpp

SST OpenCode

Google gemini-cli