// Release stream · daily archive
Fri, Oct 2
120 releases from tracked repos.
HEADLINES
HighlightGentleman-Programming/engramv3.0.0
Fri, Oct 2· 72
-
Gentleman-Programming/gentle-ai v4.0.0 Major Desktop app
Gentle AI v4.0.0 is a major release: Organic Driven Development (ODD) becomes the only development workflow as SDD and OpenSpec retire, the Go module path moves to `/v4`, OpenCode V2 gains managed plugins and native revi
-
pydantic/pydantic-ai v2.53.0 Major New agent
This release fixes one security issue in `ConcurrencyLimitedModel`. See the advisory for full details and affected versions.
-
lightpanda-io/browser 1.0.0 Major
Behavior changes
-
openclaw/openclaw v2026.8.35 Rolling New agentNew model +1 more
This is a gateway-only `extended-stable` release, which is our current equivalent to LTS. This release is OpenClaw from the end of August 2026, plus critical security updates, reliability and performance fixes, and featu
-
Gentleman-Programming/engram v3.0.0 Major +1 more
Engram 3.0.0 is a major release. It requires an explicit `expected_project` for every observation update and delete, moves the Go module path to `/v3`, adds OpenCode 2.x support, and confirms host-session registration be
-
code-yeongyu/oh-my-openagent v5.1.9 Patch Desktop app +1 more
In the desktop app's restricted modes, the agent's own tools stop asking for approval, and a bash call from an eval cell no longer hangs.** Files that ship with the app no longer count as outside paths, the first Python
-
jarrodwatts/claude-hud v0.9.0 Minor New agent +1 more
`gitStatus.showWorktree` option to show the linked worktree name after the branch, e.g. `git:(feat/x) ⎇ feat-x`, read from stdin `workspace.git_worktree` (#780).
-
router-for-me/CLIProxyAPI v8.0.9 First release +2 more
Linux release assets
-
pbakaus/impeccable skill-v4.5.0 Minor +3 more
Working screens look finished.** Dashboards, boards and tools commit to a real palette, keep the controls your users already know, and are held to the cleanest product in their category. The chosen world adds one signatu
-
Mintplex-Labs/anything-llm v1.17.0 Minor
Stop renting your intelligence. Own it with AnythingLLM. Everything you need for a powerful local-first agent experience
-
K-Dense-AI/scientific-agent-skills v2.72.0 Minor +1 more
Turn any AI agent into an AI Scientist. The #1 Agent Skills library for science, used by 250,000+ scientists worldwide. 177 ready-to-use validated skills plus 100+ scientific da…
-
sickn33/agentic-awesome-skills v18.12.0 Minor
[18.12.0] - 2026-10-02 - "Cline Orchestration, Web3 Engineering, and Safer Release Pipelines"
-
ray-project/ray ray-2.59.0 Minor
Ray Data LLM & Ray Serve LLM are GA/Stable:** the LLM APIs graduate to general availability this release (\#65194), alongside an upgrade to vLLM 0.27.0 (\#65351).
-
langfuse/langfuse v4.50.0 Minor
feat(web): show error id in tRPC error toast and include it in issue reports by @niklassemmler in https://github.com/langfuse/langfuse/pull/17427
-
github/github-mcp-server v1.13.0 Minor
GitHub's official MCP Server
-
openai/openai-python v3.24.0 Minor +1 more
3.24.0 (2026-10-02)
-
NanmiCoder/cc-haha v0.6.8 Patch Desktop app
Claude Code Haha v0.6.8
-
krillinai/OpenCreator v3.2.4 Patch Desktop app
OpenCreator v3.2.4
-
OthmanAdi/planning-with-files v3.22.0 Minor
OpenCode 2 can load `opencode-planning-with-files` 1.2.0 and use native plan injection, write reminders, compaction context, planning tools and the idle completion gate. The same package retains OpenCode 1 support.
-
czlonkowski/n8n-mcp v2.91.0 Minor
A MCP for Claude Desktop / Claude Code / Windsurf / Cursor to build n8n workflows for you
-
openai/openai-agents-python v0.23.0 Minor +1 more
A lightweight, powerful framework for multi-agent workflows
-
affaan-m/ECC v2.2.3 Patch
ECC 2.2.3 is a naming and release-reliability update. The project is called ECC
-
harvard-edge/cs249r_book tinytorch-v0.2.0 Minor
🚀 TinyTorch v0.2.0
-
triggerdotdev/trigger.dev helm-v4.7.0 Minor +5 more
Installation
-
microsoft/agent-framework python-1.20.0 Minor
[1.20.0] - 2026-10-02
-
n8n-io/n8n [email protected] Sub-package
2.41.6 (2026-10-02)
-
Niko1221/Strata v0.1.32 Patch New agent +3 more
Multi-GPU short prompts fixed, AMD decode +12%, Unsloth's Q4 model in setup with 2.8x faster prompts, and 30+ community changes - with the same answers.**
-
Arize-ai/phoenix arize-phoenix-v20.19.0 Minor
20.19.0 (2026-10-01)
-
langchain-ai/deepagents deepagents-talon==0.0.9 Release New agent +1 more
Add a Slack channel adapter, with support for outbound user mentions. (#6575, #6602)
-
DietrichGebert/ponytail v4.10.1 Patch
OpenCode 2 support. The OpenCode plugin failed to load on OpenCode 2 with "Plugin must export a default definition", and now it loads there with all commands and skills. The same entrypoint still runs on OpenCode 1 (test
-
langwatch/langwatch v3.20.0 First release +4 more
Open-Source Agent Observability & Evals for AI Platform teams. Trace, test, route and govern every agent & coding assistant in the company.
-
anthropics/claude-code v2.1.287 Patch
Added Claude Mods: plugins may now modify deeper behavior
-
langchain-ai/langchain langchain-text-splitters==1.1.3 Release
Changes since langchain-text-splitters==1.1.2
-
nanobrowser/nanobrowser v0.2.0 Minor New model
Built with WXT: the extension is now a single package built with WXT instead of a multi-package
-
paperclipai/paperclip v2026.1001.0 Rolling
Paperclip v2026.1001.0
-
ruvnet/RuView v2975 Rolling +1 more
Automated release from CI pipeline
-
cloudflare/computer @cloudflare/[email protected] Sub-package
Give your agent a computer
-
yc-software/qm v0.1.14 Patch New agent
QM 0.1.14 refines approvals and goals, adds the newest models, and gets faster under load.
-
cline/cline desktop-v0.0.42 Patch +3 more
New `/compact` slash command. It summarizes the conversation so far to free up context. The summary uses your session's provider settings and is not sent to the model as a prompt
-
googleapis/python-genai v2.27.0 Minor
2.27.0 (2026-10-01)
-
cloudflare/agents @cloudflare/[email protected] Sub-package +4 more
Build and deploy AI Agents on Cloudflare
-
vercel-labs/agent-browser v0.38.2 Patch
Browser automation CLI for AI agents
-
agno-agi/agno v3.1.1 Patch
New Features
-
volcengine/OpenViking sdk/go/v0.0.5 Patch +3 more
Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
-
sgl-project/sglang v0.5.21 Patch New model
PRs from 227 contributors.*
-
Gaurav-Gosain/tuios v0.8.5 Patch New agent
tuios 0.8.5 makes tiling more configurable and lets herdr plugins run in a tuios pane. It adds a tmux-style size policy for sessions with more than one client. Scrollback keeps its text when you resize a pane. The sideba
-
dyad-sh/dyad v1.18.0-beta.1 Minorpre
Local, open-source AI app builder for power users v0 / Lovable / Replit / Bolt alternative Star if you like it!
-
simstudioai/sim v0.9.10 Patch +1 more
feat(workflows): compare deployment versions across UI and API (#8455)
-
google/adk-java v1.11.0 Minor
An open-source, code-first Java toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control.
-
ai-dynamo/dynamo v1.5.0-motif-3-dev.1 First releasepre
A Datacenter Scale Distributed Inference Serving Framework
-
Skyvern-AI/skyvern v1.0.55 Patch
Version bumped from 1.0.56 to 1.0.55
-
earthtojake/text-to-cad v0.7.6 Patch
Other Changes
-
kortix-ai/suna v0.13.47 Patch Mobile
Microsoft Teams for every project, secrets you can share with an agent, failed-trigger alerts, and immediate sign-out
-
confident-ai/deepeval python-v4.2.8 Patch
The LLM Evaluation Framework
-
coder/coder v2.34.12 Patch
fix(scripts/releaser): remove GPG prompt and scope breaking-change check (#27422, #29487) (#29498, 4f9a539651)
-
YishenTu/claudian 2.3.12 Patch
feat(chat): show composer commands and skills as chips (#1508)
-
livekit/agents [email protected] Sub-package
feat(smallestai): add TTS continuations protocol support by @harshitajain165 in https://github.com/livekit/agents/pull/7111
-
run-llama/liteparse node-v2.15.1 Patch +3 more
fix: is_garbled checks for broken fonts by @logan-markewich in https://github.com/run-llama/liteparse/pull/479
-
google-gemini/gemini-cli v0.64.0-nightly.20261002.gc9096a847 Patchpre
fix(core): implement append-only delta patching and bounded history windowing in ChatRecordingService by @jvargassanchez-dot in https://github.com/google-gemini/gemini-cli/pull/29568
-
mobile-next/mobile-mcp 1.0.8 Patch +1 more
Model Context Protocol Server for Mobile Automation and Scraping (iOS, Android, Emulators, Simulators and Real Devices)
-
Lightricks/LTX-2 v1.4.2 Patch
Official Python inference and LoRA trainer package for the LTX-2 audio–video generative model.
-
zhouxiaoka/autoclip v1.5.1 Patch
AutoClip|一个链接,一键出片。开源 AI 视频剪辑桌面工具,将播客、访谈、课程等长视频自动剪成短视频,生成字幕、封面和发布文案,适配抖音、小红书、TikTok、Reels 与 YouTube Shorts。Open-source AI video clipping & content repurposing.
-
strands-agents/harness-sdk python/v1.57.2 Sub-package
python/v1.57.2
-
mvschwarz/openrig v0.6.4 Patch +1 more
OpenRig 0.6.4
-
Giskard-AI/giskard-oss v3.0.1 Patch +5 more
feat(checks): accept model strings in set_default_generator by @henchaves in https://github.com/Giskard-AI/giskard-oss/pull/2794
-
microsoft/mcp Azure.Mcp.Server-3.0.0-beta.49 Patch IDE extension
_ADME, Azure Backup Vault and Local Auth_
-
UKGovernmentBEIS/inspect_ai 0.3.276 Patch +1 more
Inspect: A framework for large language model evaluations
-
Comfy-Org/ComfyUI_frontend v1.54.11 Patch
[backport core/1.54] test: restore abandoned-slice browser coverage (#18840) by @christian-byrne in https://github.com/Comfy-Org/ComfyUI_frontend/pull/19700
-
anthropics/claude-agent-sdk-typescript v0.3.287 Patch
Added optional remote-session latency fields (`first_text_post_queue_wait_ms`, `first_text_post_queued_behind`) to the success result message
-
ibm-granite/granite-tsfm v0.3.10 Patch
Granite tsfm ensemble by @fearghalodonncha in https://github.com/ibm-granite/granite-tsfm/pull/546
-
flashinfer-ai/flashinfer v0.7.1rc2 Patchpre
FlashInfer: Kernel Library for LLM Serving
-
gptme/gptme v0.34.1.dev20261001 Patchpre
Your agent in your terminal, equipped with local tools: writes code, uses the terminal, browses the web. Make your own persistent autonomous agent on top!
Nothing matches this filter.