// Release stream · daily archive
Fri, Oct 16
205 releases from tracked repos.
Fri, Oct 16· 123
-
mastra-ai/mastra @mastra/[email protected] Sub-package New agent
Agent Scopes (multi-tenant identity + safer memory/thread routing)
-
DietrichGebert/ponytail v5.3.0 Minor +1 more
Ponytail 5.3 puts a band above the Claude Code prompt: what the current task costs, its tokens and agent time, how much existing code it reused, and whether the tests still pass. One click opens the board, a local page w
-
huggingface/transformers v5.20.0 Minor New model +1 more
Release v5.20.0
-
skyhook-io/radar v1.17.0 Minor Desktop app
⚠️ Security fix — upgrade recommended
-
coreyhaines31/marketingskills v2.11.35 First release +2 more
Marketing skills for Claude Code and AI agents. CRO, copywriting, SEO, analytics, and growth engineering.
-
thedotmack/claude-mem v13.40.0 Minor
Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future se…
-
koala73/worldmonitor v2.11.0 Minor
chore: bump version 2.10.0 → 2.11.0 (#9182)
-
Comfy-Org/ComfyUI v0.41.0 Minor New model
Add ImageSDRToHDR node. by @comfyanonymous in https://github.com/Comfy-Org/ComfyUI/pull/16990
-
pbakaus/impeccable cli-v4.6.0 Minor +1 more
Install works on Node 18.19 and later.** The package no longer asks for Node 22.18, so npm stops falling back to the old CLI.
-
openclaw/openclaw v2026.10.10-beta.1 Rollingpre Desktop appNew model +1 more
Faster, steadier Gateway:** chat turns, session lists, startup, and idle Gateways do far less SQLite and main-thread work, with more database work moved onto worker threads, so busy hosts and large agent fleets stay resp
-
career-ops-hq/career-ops career-ops-v1.37.0 Minor +1 more
This release is about knowing what is still worth your time: postings you are tracking get checked for you, and every report opens with its verdict. The dashboard work (interview prep, cover-letter drafts, row comparison
-
docling-project/docling v2.140.0 Minor
Get your documents ready for gen AI
-
anthropics/claude-code v2.1.301 Patch New agentNew model
Added `CLAUDE_CODE_RETRY_WATCHDOG_MAX_USAGE_LIMIT_WAIT_MS` so unattended retry mode (`CLAUDE_CODE_RETRY_WATCHDOG`) ends a request at a usage limit that resets too far away instead of waiting
-
twentyhq/twenty sdk/v2.48.0 Sub-package +1 more
⚠️ Breaking changes
-
sickn33/agentic-awesome-skills v19.8.0 Minor
[19.8.0] - 2026-10-16 - "Fifteen Reviewed Additions And A Healthier Review Lane"
-
stablyai/orca v1.4.227 Patch New agent
Thank you so much for using Orca and for your continued support! ❤️
-
upstash/context7 [email protected] Sub-package +3 more
Context7 Platform -- Up-to-date docs for AI agents
-
CopilotKit/CopilotKit intelligence-mastra/v1.72.0 Sub-package +4 more
feat: consume learned skills from multiple containers across SDKs (#7384) (6252d10)
-
ollama/ollama v0.40.6-rc0 First releasepre
Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
-
langfuse/langfuse v4.63.0 Minor +2 more
feat(web): link personal feature previews from organization settings by @nimarb in https://github.com/langfuse/langfuse/pull/18887
-
NanmiCoder/cc-haha v0.7.4 Patch Desktop app +1 more
Claude Code Haha v0.7.4
-
DeusData/codebase-memory-mcp v0.12.0 Minor
Local code intelligence for developers and AI agents. Connect code, docs and infrastructure in a persistent graph, with CBM Lens, MCP tools, agent integrations and change-aware…
-
krillinai/OpenCreator v3.2.7 Patch Desktop app
OpenCreator v3.2.7
-
tashfeenahmed/freellmapi v0.15.0 Minor
7.4 billion tokens per month. 34 free LLM providers. 635 free model endpoints. All behind one /v1 endpoint, plus any custom OpenAI-compatible endpoint. Smart routing, automatic…
-
Imbad0202/academic-research-skills v3.23.1 Patch New agent
Repairs and model currency; no new skill, and no accuracy or performance claim:** v3.23.1 closes two write-scope guard paths that let writes through without a check, one for plugin-dispatched agents and one for Windows `
-
n8n-io/n8n [email protected] Sub-packagepre +1 more
2.44.3 (2026-10-16)
-
pydantic/pydantic-ai v2.57.0 Minor
⚠️ Compatibility Notes
-
huggingface/trl v1.16.0 Minor New model
🛠️ TRL now trains on AWS Trainium
-
confident-ai/deepeval python-v4.3.0 Minor
The LLM Evaluation Framework
-
citrolabs/ego-lite v2.1.0 Minor
Other Changes
-
web-infra-dev/midscene v1.16.0 Minor +2 more
See interfaces like a human. E2E testing in natural language.
-
can1357/oh-my-pi v18.11.2 Patch New agent
@oh-my-pi/pi-agent-core
-
anomalyco/opencode v1.19.3 Patch +2 more
Improvements
-
akitaonrails/ai-memory v3.1.0 Minor +1 more
Solution for long term memory for agent coding CLIs and to facilitate handoff between different agent vendors
-
Arize-ai/phoenix arize-phoenix-v20.23.0 Minor
20.23.0 (2026-10-16)
-
langchain-ai/deepagents deepagents==0.7.26 Patch New agent
Async subagents now report when they’re waiting for a human. (#6842)
-
QwenLM/qwen-code sdk-typescript-v0.1.20 Patch New agent +2 more
Bundled CLI Version
-
Canner/WrenAI wren-v0.16.0 Minor
0.16.0 (2026-10-16)
-
Kilo-Org/kilocode jetbrains/v7.1.10 Sub-package New agent
Show and edit agent models from the agents list.
-
vitali87/code-graph-rag v0.2.5 First release +1 more
The ultimate RAG for your monorepo. Query, understand, and edit multi-language codebases with the power of AI and knowledge graphs
-
open-webui/open-webui v0.12.3 Patch
🔐 **End-to-end encrypted chats.** Direct messages and private group channels can now hold a temporary end-to-end encrypted chat: a member picks Encrypted Chat in the channel menu, the other members see "Encrypted chat in
-
farion1231/cc-switch v4.0.14 Patch +1 more
CC Switch v4.0.14
-
langchain-ai/langchain langchain-core==1.6.11 Patch
Changes since langchain-core==1.6.10
-
axolotl-ai-cloud/axolotl v0.21.0 Minor
Go ahead and axolotl questions
-
Gentleman-Programming/gentle-ai v4.2.0 Minor
Gentle AI v4.2.0 is a minor release. It publishes Gentle Shell's ODD routing block from gentle-ai, so Pi's workflow wording now has a single source, and it ships the orchestrator routing changes measured on the Gentle Sh
-
browser-use/browser-use 0.13.13 Patch +1 more
Open-source Python library for AI browser agents, plus Browser Use Cloud: stealth browsers and a hosted agent API.
-
google-gemini/gemini-cli v0.66.0-nightly.20261016.gdebf42324 Minorpre
chore(release): bump version to 0.66.0-nightly.20261013.g204db94ae by @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/29740
-
vercel/eve eve%400.80.0 Minor +2 more
The Open Framework for Building Agents
-
tangency-labs/solidcraft v0.14.0 Minor +3 more
SolidCraft is open-source parametric 3D CAD in Rust: sketch, constrain, extrude, and change a parameter to watch the timeline rebuild. Documentation · try it in the browser · contributing · issues.
-
unslothai/unsloth v0.1.908-beta Patch
Run NVIDIA's new Nemotron 3.5 models and Meta's new Muse Spark Open locally in Unsloth. New: text-diffusion LLMs, DeepSpeed training and better prompt caching for coding agents.
-
alchaincyf/huashu-art-motion v1.2.0 Minor
新风格：敦煌壁画（北朝）。** 土红地与奶白地两栏、白鹿与锯齿山、横飞的飞天和青绿风叶，墙面有龟裂、起甲和盐碱。
-
ml-explore/mlx-lm v0.32.0 Minor +2 more
Run LLMs with MLX
-
code-yeongyu/oh-my-openagent v5.1.34 Patch
`oracle`, `momus` and `plan-reviewer` fall back to Argon before the previous pro model.** In all three chains, Argon at high thinking (Google, Vertex or OpenCode) now sits directly ahead of the previous pro rung, which s
-
cline/cline desktop-v0.0.47 Patch +3 more
Marketplace plugins such as goal load again. In 0.0.46 they failed with `Cannot find module '@cline/sdk'` because the packaged app didn't include the SDK packages plugins import
-
heygen-com/hyperframes v0.8.164 Patch +1 more
HyperFrames v0.8.164
-
haris-musa/excel-mcp-server v2.1.0 Minor +1 more
PivotTables over large sources (up to 3 million cells by default, configurable), faster pivot creation, new limit options, and complete nested parameter docs in TOOLS.md.
-
hugohe3/ppt-master v6.7.1 Patch
Patch release: fixes for work that the live preview, source conversion and source import could lose without a warning, plus a full pass over the UI strings in every language.
-
ultralytics/ultralytics v8.4.178 Patch
Ultralytics YOLO27, YOLO26, YOLO11, YOLOv8 — object detection, instance segmentation, semantic segmentation, image classification, pose estimation, object tracking
-
anthropics/anthropic-sdk-python v1.15.0 Minor
api:** add between_tools to thinking types in model capabilities
-
modelcontextprotocol/java-sdk v2.2.0-M1 Minor +3 more
The official Java SDK for Model Context Protocol servers and clients. Maintained in collaboration with Spring AI
-
router-for-me/CLIProxyAPI v8.0.31 Patch
Linux release assets
-
strands-agents/harness-sdk python/v1.61.1 Sub-package New agent +1 more
python/v1.61.1
-
Gaurav-Gosain/tuios v0.10.0 Minor
tuios 0.10.0 is a feature release. On Omarchy, tuios now uses the Omarchy theme when you set no theme, and it changes its colours when you switch the Omarchy theme. You do not restart tuios.
-
multica-ai/multica v0.6.3 Patch
c059537cd60daf11d152bd76a2e3fc2a9f MUL-6945 & MUL-6944: fix(daemon): preserve Git content during artifact GC (#7893)
-
yc-software/qm desktop-v0.1.3 Patchpre Desktop app
QM Desktop 0.1.3 for Apple Silicon Macs adds local computer access through the `desktop` tool. Human messages sent from the desktop app can observe windows, take screenshots, click and type, with guided Accessibility and
-
gptme/gptme v0.35.0 Minor +2 more
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!
-
agno-agi/agno v3.1.4 Patch
New Features:
-
Comfy-Org/workflow_templates v0.11.85 First release +1 more
ComfyUI template workflows
-
VectifyAI/PageIndex v0.4.1 Patch
PageIndex: Document Index for Vectorless, Reasoning-based RAG
-
Tencent-Hunyuan/Hy-Memory v1.0.0 First release
Hy Memory and MemHub v1.0.0.
-
dyad-sh/dyad v1.20.0-beta.1 Minorpre
Local, open-source AI app builder for power users v0 / Lovable / Replit / Bolt alternative Star if you like it!
-
onyx-dot-app/onyx helm/onyx-0.9.16 Sub-package +2 more
Onyx application and Craft sandbox deployment chart
-
topoteretes/cognee v1.6.5.dev0 Patch
v1.6.5.dev0 — Distillation, Embeddings & Connector Hardening
-
trycua/cua cua-driver-rs-v0.36.0 Minorpre +6 more
Cua Driver 0.36.0
-
simstudioai/sim v0.9.24 Patch
feat(memory): authorize scoped note access and durable cleanup (#9096)
-
Infisical/infisical v0.166.8 Patch
feat: add semantic Secrets Overview telemetry by @andrewhuhh in https://github.com/Infisical/infisical/pull/8636
-
manaflow-ai/cmux v0.65.2 Patch +1 more
ci: run backend migrations gate on every PR again (revert #17415) by @teamleaderleo in https://github.com/manaflow-ai/cmux/pull/17432
-
OthmanAdi/planning-with-files v3.24.1 Patch
Frontmatter validation now covers all eighteen tracked skills, including the twelve hidden adapter copies. Isolated fixtures check hidden valid and malformed YAML, excluded directories and Windows path aliases. A tracked
-
deepseek-ai/deepseek-harness dsh-v0.2.1-alpha.5 Patchpre
DeepSeek Harness: Everything is a Plugin.
-
NVIDIA/NemoClaw v0.0.135 Patch
Run agents like Hermes, LangChain Deep Agents, and OpenClaw more securely inside NVIDIA OpenShell with managed inference
-
langchain-ai/langchainjs @langchain/[email protected] Sub-package +4 more
The agent engineering platform
-
alibaba/zvec v0.8.0-alpha Minorpre
feat(tools): support turbo quantizer in core tool set by @richyreachy in https://github.com/alibaba/zvec/pull/684
-
YishenTu/claudian 2.3.22 Patch
fix: keep chat history readable across provider runtime changes (#1647)
-
eigent-ai/eigent v1.0.6 Patch
Eigent: The Open Source Cowork Desktop - Local and Free Alternative to Claude Cowork and Codex
-
modelcontextprotocol/ruby-sdk v1.8.0 Minor
The official Ruby SDK for the Model Context Protocol servers and clients.
-
ai-dynamo/dynamo v1.6.0-nemotron-3.5-super-vl-dev.1 Minorpre
A Datacenter Scale Distributed Inference Serving Framework
-
apify/apify-mcp-server v0.18.4 Patch
0.18.4 (2026-10-16)
-
LMCache/LMCache v0.5.6 Patch
LMCache: Supercharge Your LLM with the Fastest KV Cache Layer
-
langchain-ai/open-swe desktop-v0.3.8 Patch +6 more
fix(ui): highlight review diff hunks with full file context by @johannes117 in https://github.com/langchain-ai/open-swe/pull/4074
-
awslabs/mcp 2026.10.20261015170211 Rolling
feat(redshift-mcp-server): opt-in read-write mode and named transactions by @grayhemp in https://github.com/awslabs/mcp/pull/4627
-
xorbitsai/inference v3.5.0 Patch +1 more
Swap GPT for any LLM by changing a single line of code. Xinference lets you run open-source, speech, and multimodal models on cloud, on-prem, or your laptop — all through one un…
-
MervinPraison/PraisonAI v4.7.15 Patch
PraisonAI — Hire a 24/7 AI Workforce. Stop writing boilerplate and start shipping autonomous self-improving agents that research, plan, code, and execute tasks. Deployed in 5 li…
-
paradedb/paradedb v0.26.2 Patch
fix: install ICU headers in publish builds by @paradedb-github-bot[bot] in https://github.com/paradedb/paradedb/pull/6756
-
rtk-ai/rtk dev-1.0.2-rc.549 Patchpre
CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single Rust binary, zero dependencies
-
anthropics/claude-agent-sdk-python v0.2.168 Patch
Internal/Other Changes
-
lobehub/lobehub v0.0.0-nightly.pr20458.35130 Patchpre +2 more
LobeHub is your Chief Agent Operator, organizing your agents into 7×24 operations by hiring, scheduling, and reporting on your entire AI team.
-
leejet/stable-diffusion.cpp master-968-8cd731e Rolling +2 more
Diffusion model(SD,Flux,Wan,Qwen Image,Z-Image,...) inference in pure C/C++
-
vastsa/PI-Desktop v0.21.0-beta.1 Minorpre
Local-first AI coding agent desktop: Electron + Rust host core + pi Agent Harness + user-installable plugins
-
archestra-ai/archestra platform-v1.4.0-rc.43 Minorpre +5 more
1.4.0-rc.43 (2026-10-15)
-
flashinfer-ai/flashinfer v0.7.2 Patch
_These highlights are also published at flashinfer.ai/releases._
-
truefoundry/trueforge @truefoundry/[email protected] Sub-package
The open-source agent harness - the runtime layer that turns an LLM into a working agent.
-
netease-youdao/LobsterAI 2026.10.16 Rolling
✨ Highlights
-
OpenBMB/PilotDeck v2026.10.16 Rolling
Task-oriented AI Agent productivity platform
-
Agenta-AI/agenta v0.122.7 Patch
fix(frontend): open the billing portal and checkout for the selected organization by @mmabrouk in https://github.com/Agenta-AI/agenta/pull/7619
-
siyuan-note/siyuan v3.8.8-alpha.2 Patchpre +1 more
Upgrade to Electron v44.8.0
-
zhongerxin/iPhone-use v0.3.15 Patch +1 more
新增 `pua_screenshot()`：通过 WDA HID 事件直接触发 iOS 系统截屏，图片保存到 iPhone 相册。需要在手机里发送、上传或使用截图时，调用工具后到「照片」App 或目标 App 的相册选择器中找到并使用。
-
QingYunA/answer-me-with-html v0.5.5 Patch
Write the numbers in a `chart` block and the page draws the chart (#264, #251):
-
langroid/langroid 0.69.8 Patch
Harness LLMs with Multi-Agent Programming
-
microsoft/mcp Template.Mcp.Server-0.0.12-alpha.6964643 Patch
Catalog of official Microsoft MCP (Model Context Protocol) server implementations for AI-powered data access and tool integration
-
samuelgursky/davinci-resolve-mcp v4.13.2 Patch
v4.13.2 — Three more Windows test failures fixed
-
lmnr-ai/lmnr v0.3.2 Patch
Laminar - open-source observability platform purpose-built for AI agents. YC S24.
-
UKGovernmentBEIS/inspect_ai 0.3.281 Patch
Inspect: A framework for large language model evaluations
-
ollama/ollama-vscode v0.0.12 First release +1 more
Official Ollama Extension for VS Code
-
ROCm/FastFlowLM v1.0.9 Patch
This is a maintenance release. It fixes **`flm validate` on Linux**, which wrongly reported "No NPU device found." in v1.0.8.
-
anthropics/claude-agent-sdk-typescript v0.3.301 Patch
Added per-tool permission policies to sdk-type MCP server configs: `tools` entries with `permission_policy` `always_ask` or `always_deny`, applied on `initialize` and `setMcpServers()`
-
intel/auto-round v0.16.1 Patch
PTQ-Level Efficiency, QAT-Level Accuracy|训练后量化的效率，量化感知训练的精度
-
lancedb/lancedb v0.41.0-beta.7 Patchpre
🛠 Breaking Changes
-
google-ai-edge/LiteRT-LM 0.19.0-nightly.20261015 Patchpre
LiteRT-LM is Google's production-ready, high-performance, open-source inference framework for deploying Large Language Models on edge devices.
-
gpustack/gpustack v2.3.0rc3 Patchpre
A GPU cluster manager for high-performance AI model serving (vLLM, SGLang) and on-demand SSH-accessible GPU instances.
-
apache/maka v0.2.1-dev.88.20261016 Patchpre
Developer Snapshot 0.2.1-dev.88.20261016
-
Tencent/teamai-cli v0.28.0-beta.5 Patchpre
Make Every Team AI Native
-
awslabs/aidlc-workflows v2.11.1-preview.20261015.1 Patchpre
AI-Driven Life Cycle (AI-DLC) adaptive workflow steering rules for AI coding agents
-
modal-labs/modal-client py%2Fv1.6.2 Patch +2 more
SDK libraries for Modal
Nothing matches this filter.