DevHunt Daily
Dev tools & AI news, every day
Tuesday, November 17, 2026
- 1SecurityHacker News
Optimizely snippets compromised, injecting malware into sites including Zoom and GitLab
Malicious JavaScript was added to Optimizely Web snippets for 178 organizations, delivering crypto‑stealing and card‑skimming code. The attack affected high‑traffic sites such as Zoom, GitLab, Salesforce and Smartsheet, highlighting risks of third‑party script dependencies.
- 2SecurityLobsters
Universal XSS vulnerability discovered and patched in uBlock Origin's element picker
A one‑click universal XSS flaw was found in uBlock Origin's element picker, allowing a malicious site to add trusted filters that execute arbitrary JavaScript on any page. The issue has been fixed in the latest extension release, and details are available in the GitHub security advisory.
- 3Dev toolsGitHub Changelog
Dependabot delegated alert dismissals now stay open for one year
GitHub extended the lifespan of delegated Dependabot alert dismissal requests from one week to one year, giving reviewers more time to act. The change is live on github.com and will be included in GitHub Enterprise Server 3.24.
github.blog - 4AIGitHub Changelog
GitHub Copilot deprecates several AI models on November 16, 2026
GitHub Copilot has retired the Claude Opus 4.8, Claude Opus 5, and GPT-5.6 Terra models as of November 16, 2026. Users should switch to the suggested alternatives—Claude Opus 5.5 or GPT-6.1 Sol/GPT-6 Luna—through Copilot settings. The change affects all Copilot experiences, including chat and code completions.
github.blog - 5AIVercel
Vercel AI Gateway adds Private Inference for self‑hosted models
AI Gateway now lets you connect your own model endpoints as private providers, using the same authentication, SDKs, and observability as catalog models. Developers can switch between private and public models, keep them team‑scoped, and monitor traffic through Vercel’s logs and tracing.
vercel.com - 6AIHugging Face
Hugging Face switches inference routing to OpenRouter for open-weight models
Hugging Face now routes all inference requests for open-weight models through OpenRouter, which selects the best provider based on price, speed and reliability. The change is transparent to developers; existing code and API calls remain unchanged while benefiting from automatic provider selection and retries.
huggingface.co - 7AIGitHub Trending
Univer releases open-source Office SDK for building custom spreadsheet and document editors
Univer Office SDK provides editable sheets, docs, and other office features that can be embedded in applications and used by AI agents. The SDK is open source, supports collaboration and file processing, and includes a reference workspace project.
github.com1,828 stars today - 8AIGitHub Trending
NVIDIA releases SkillEvaluator for testing AI agent skills and plugins
SkillEvaluator is a Python framework that checks AI agent skills for quality, security, overlap and live impact. It can run from the command line, CI pipelines or be integrated into agents, helping developers validate and improve their skill packages.
github.com873 stars today - 9AIGitHub Trending
PaperOrchestra launches multi‑agent framework for automated AI research paper writing
Google Research released PaperOrchestra, a framework that coordinates specialized agents to turn raw research notes into LaTeX manuscripts with literature reviews and generated figures. Developers can use it to automate paper drafting and integrate with Gemini or other LLM APIs.
github.com313 stars today - 10Dev toolsGitHub Trending
Plannotator adds browser‑based inbox for reviewing AI agent output
Plannotator introduces a local, browser‑based Inbox that aggregates messages, plans, and code reviews from multiple AI coding agents. Developers can annotate, comment, and send feedback directly back to the agents, streamlining multi‑agent workflows.
github.com696 stars today - 11AISimon Willison
LLM 0.37.3 fixes Homebrew SQLite logging issue
Version 0.37.3 of the LLM command‑line tool addresses a breakage caused by Homebrew upgrading to SQLite 3.54.1, which changed JSON handling in queries. The release updates the logging code to work with the corrected SQLite behavior, preventing errors for developers using LLM with Homebrew.
simonwillison.net - 12Open sourceSimon Willison
Datasette 1.0a43 adds streaming output, security fixes and CSS view transitions
The new release introduces streaming support for output renderers, enabling TSV and CSV exports to stream rows. It also includes security patches for private stored queries and adds CSS multi‑page view transitions to improve navigation feel. Developers using Datasette will benefit from faster exports and enhanced security.
simonwillison.net
Monday, November 16, 2026
- 1AIHacker News
Qwen releases Qwen4-Max with 2.4T parameters and 1M-token context
Qwen4-Max, the latest model from Qwen, features a 2.4‑trillion‑parameter backbone, native vision support and a one‑million‑token context window. The model improves token efficiency, cutting costs by up to 62% on benchmark tasks, and shows better performance across coding, work and long‑horizon tasks, which may benefit developers building AI‑assisted applications.
- 2WebHacker News
Protomaps PMTiles enable cheap self‑hosted map tiles
Protomaps' PMTiles format stores map tiles in a single file served via HTTP Range requests, making deployment to CDNs simple and inexpensive. Using a storage bucket like Bunny.net can reduce costs to a few dollars per month compared to commercial APIs. Developers can embed the tiles with MapLibre GL after installing the pmtiles npm package.
- 3SecurityHacker News
Apollo missions carried grenade launchers for seismic experiments
NASA equipped some Apollo landings with a grenade launcher and grenades to create controlled surface explosions. The resulting shockwaves helped scientists study lunar subsurface structure through seismic imaging, providing data on how seismic waves travel underground.
- 4SecurityHacker News
Nvidia board faces lobbying to reinstate CEO Jensen Huang after his resignation
Jensen Huang resigned from Nvidia following allegations of an affair with a suspected Chinese spy. Silicon Valley allies are organizing a lobbying effort, including a “war room,” to persuade the board and possibly the White House to bring him back. Developers may watch for potential impacts on Nvidia’s strategy and product roadmaps.
- 5SecurityHacker News
Tor releases 0.4.9.17 with DoS defense fix amid ongoing DDoS attack
The Tor Project published security release 0.4.9.17, adding several fixes including an important mitigation for relay denial‑of‑service attacks related to IPv6 handling. Relay operators are urged to upgrade immediately to help protect the network from the current large‑scale DDoS campaign.
- 6Open sourceHacker News
Eclipse Mosquitto repository receives recent CMake install target fix
The Mosquitto MQTT broker project on GitHub shows ongoing maintenance, with the latest commit fixing the CMake install target. Developers using Mosquitto can expect continued updates and bug fixes to improve build reliability.
- 7Coding agentsGitHub Trending
AG-UI launches open protocol for integrating AI agents with frontend apps
AG-UI is an open, lightweight, event‑based protocol that standardizes how AI agents communicate with user‑facing applications. It supports real‑time chat, state sync, and works with transports like SSE and WebSockets, helping developers embed agents in web interfaces.
github.com355 stars today - 8Coding agentsGitHub Trending
Google releases Stitch Skills library compatible with coding agents
Google published a collection of Agent Skills for the Stitch MCP server, following the Agent Skills open standard. The skills can be installed via Codex, Claude Code, Cursor or OpenCode and enable design, build and utility workflows for agents. Developers can add the plugins to extend AI coding agents with design generation and management capabilities.
github.com247 stars today - 9WebGitHub Trending
Google Chrome releases Modern Web Guidance source repository
The Modern Web Guidance project provides a set of curated skills that embed up‑to‑date web platform best practices into coding agents. It offers a CLI for installing, searching, and scanning code to replace legacy patterns with modern APIs, improving performance and accessibility for developers.
github.com1,219 stars today - 10SecurityGitHub Trending
Claude-BugHunter adds 83 bug‑hunting skills for Claude Code and other AI agents
Claude-BugHunter is a skill bundle that adds 83 bug‑hunting and red‑team skills, 15 slash commands, and hundreds of report patterns to Claude Code and compatible AI agents. Developers can install it via the Claude Code plugin or a manual install to automate vulnerability research workflows.
github.com98 stars today - 11SecuritySimon Willison
Datasette 0.65.6 backports security fix for route-based database download
Datasette 0.65.6 includes a backported security patch from version 1.0a43. The fix prevents unauthorized downloading of databases when a custom route differs from the database name. Developers using add_database with custom routes should upgrade to avoid this exposure.
simonwillison.net - 12LanguagesLobsters
clice 1.0 released as a modern C++ language server
clice 1.0, a C/C++ language server built with C++20/23 features, is now stable after extensive testing. It offers process‑isolated compilation, module support, and a hermetic Bazel build, aiming to improve reliability over clangd and ccls.
- 13AIGitHub Trending
Claude Trading Skills toolkit offers AI‑assisted workflows for individual investors
The open‑source Claude Trading Skills project provides a set of Claude‑based workflows to help time‑constrained investors structure market review, risk management, trade planning and journaling. It targets long‑term, ETF and dividend investors who also use disciplined swing‑trading, without automating buy‑sell decisions.
github.com706 stars today
Sunday, November 15, 2026
- 1AIHacker News
DeepSeek V3 implemented in ~200 lines of pure JAX
A GitHub repo provides a compact JAX implementation of the DeepSeek V3 model, covering model definition, training, sampling, data preparation, and tests in roughly 200 lines. Developers can use it to experiment with the open-weight DeepSeek model without extensive codebases.
- 2AIHacker News
Serving GLM-5.3 across two NVIDIA DGX stations with vLLM
NVIDIA demonstrates how to run the 753‑billion‑parameter GLM-5.3 model on two DGX stations using tensor parallelism and vLLM, achieving up to 607 tokens per second. Developers can follow the guide to deploy large models that exceed a single GPU's memory.
- 3SecurityHacker News
Project Zero shifts to a 90‑day disclosure timeline, plans 60‑day then 30‑day limits
Google's Project Zero announced it will disclose reported vulnerabilities after 90 days, moving to 60 days in early 2027 and 30 days in 2028. The change aims to speed up vendor remediation and reduce the gap between patch release and user installation as AI accelerates exploit development.
- 4Dev toolsGitHub Trending
Alibaba releases OpenCodeReview, an AI-powered CLI for precise code reviews
OpenCodeReview is a command‑line tool that uses LLM agents to analyze Git diffs and add line‑level comments, supporting multiple languages and custom rules. It claims higher precision and lower token usage than generic agents, aiming to reduce false positives in CI pipelines.
github.com893 stars today - 5Dev toolsGitHub Trending
Uptime Kuma offers self‑hosted monitoring with Docker and native installs
Uptime Kuma is an open‑source, self‑hosted tool for monitoring website, service and container uptime. It provides a web UI, multiple notification channels and supports Docker, Docker‑Compose and manual Node.js installations, which is useful for developers needing on‑premise health checks.
github.com430 stars today - 6AIGitHub Changelog
GitHub Copilot app adds general availability for GitHub Enterprise Server support
The Copilot app (v1.2.0+) now lets users add a GitHub Enterprise Server account to work with remote repositories, issues, and pull requests. Cloud sessions remain unavailable for GHES. Developers using self‑hosted GitHub can now access Copilot features directly from the app.
github.blog - 7CloudVercel
Vercel Daemons can now send OpenTelemetry traces to Vercel Tracing
Vercel added support for daemons to export OpenTelemetry traces via a Unix socket, using @vercel/otel 2.3.0+. Developers can view daemon spans in the Vercel CLI, enabling tracing of background workers alongside other project traces.
vercel.com - 8AIVercel
Vercel Token Pass raises monthly token credit to $150
Vercel’s Token Pass add‑on now provides $150 of AI Gateway tokens each month, up from $80, while keeping the price at $30 per team. The credit covers open‑weight models like Kimi, GLM, and DeepSeek, and resets each billing cycle with no rollover. Developers can enable it via the Vercel CLI or API key.
vercel.com - 9SecurityCloudflare
Cloudflare warns of rapid rise in AI-driven penetration testing frameworks
Cloudflare reports a surge in open‑source LLM‑based pentesting tools like ARTEX and its many forks, with thousands of repositories and a 30× increase in daily requests. The firm advises adopting a Modern Security Model with WAF, bot defense, identity‑gated access and threat intelligence to mitigate these autonomous attacks.
blog.cloudflare.com - 10DataSimon Willison
datasette-llm adds OpenAI‑compatible endpoint support
Version 0.1a11 of the datasette-llm plugin adds support for OpenAI‑compatible endpoints, allowing any LLM that follows the OpenAI web protocol to be configured with Datasette. This enables developers to run local models from LM Studio, Ollama or llama.cpp with Datasette agents via uvx.
simonwillison.net - 11AIProduct Hunt
CodeMesh reduces AI coding token usage by up to 95%
CodeMesh creates a live map of a codebase and supplies AI coding agents only the relevant parts, cutting token consumption and API costs dramatically. In tests on a real repository it lowered token use by 95% and costs by over 90%, benefiting developers using Claude Code, Cursor or Windsurf.
producthunt.com - 12SecurityProduct Hunt
Antelog launches to monitor and block risky AI coding agent actions
Antelog, a free tool, records every command, website visit, and data transfer made by AI coding agents and blocks potentially destructive actions such as deleting databases or exposing secrets. It integrates with Claude Code, Cursor, Codex and similar agents, giving developers visibility and safety when using AI assistants.
producthunt.com - 13AIDevHunt launch
ForAgent adds private phone numbers for AI agents to call each other
ForAgent provides private phone numbers that let AI agents communicate directly, either on local machines or with external contacts. Calls can be handled via the app or a plugin, supporting agents like Claude Code, Codex, Grok, Cursor, and others. The free tier offers 30 external calls per month, with internal calls always free.
14 points - 14LanguagesDevHunt launch
Rustable launches as a fast, lightweight native database client written in Rust
Rustable is an open-source database client built in Rust, positioned as a lightweight alternative to TablePlus. Its native implementation aims to provide speed and functionality for everyday database tasks, which may interest developers seeking performance-focused tools.
10 points
Archive
- Saturday, November 14, 2026 · 14 stories
- Friday, November 13, 2026 · 13 stories
- Thursday, November 12, 2026 · 13 stories
- Wednesday, November 11, 2026 · 11 stories
- Tuesday, November 10, 2026 · 14 stories
- Monday, November 9, 2026 · 12 stories
- Sunday, November 8, 2026 · 15 stories
- Saturday, November 7, 2026 · 12 stories
- Friday, November 6, 2026 · 12 stories
- Thursday, November 5, 2026 · 15 stories
- Wednesday, November 4, 2026 · 14 stories
- Tuesday, November 3, 2026 · 14 stories
- Monday, November 2, 2026 · 12 stories
- Sunday, November 1, 2026 · 15 stories
- Saturday, October 31, 2026 · 12 stories
- Friday, October 30, 2026 · 12 stories
- Thursday, October 29, 2026 · 15 stories
- Wednesday, October 28, 2026 · 14 stories
- Tuesday, October 27, 2026 · 15 stories
- Monday, October 26, 2026 · 12 stories
- Sunday, October 25, 2026 · 13 stories
- Saturday, October 24, 2026 · 14 stories
- Friday, October 23, 2026 · 10 stories
- Thursday, October 22, 2026 · 13 stories
Built something developers should hear about? Launch it on DevHunt: the week’s top launches make the news here.