2/17/2026
Show HN: Twick – React Video Editor SDK with AI Captions and MP4 Export
React video editor SDK with AI-powered caption generation, timeline editing, and serverless MP4 export for building video SaaS applications.
2/17/2026
Show HN: Agntor – Trust infrastructure for AI agents (identity, escrow, guard)
Trust infrastructure library for AI agents providing identity verification, escrow settlement, reputation, and multi-layer prompt injection detection. Node.js package.
2/17/2026
Show HN: Stop paying the $1,000 "Amnesia Tax" on your AI agents
Solution addressing memory/context limitations in AI agents, framed as reducing operational costs.
2/17/2026
Show HN: Agentsec – Security scanner for AI agent installations (MCP, OpenClaw)
Security scanner tool for AI agent installations, supporting MCP and OpenClaw platforms.
2/17/2026
Show HN: A vision-based AI agent for end-to-end testing
Vision-based AI agent for end-to-end testing using natural language and visual recognition, eliminating need for test automation code.
2/17/2026
Ask HN: What's stopping you from running autonomous agents today?
Discussion of practical challenges for deploying autonomous agents on commodity infrastructure: security, monitoring, cost tracking, API management, scaling.
2/17/2026
Why does GPT-5.1 Codex underperform GPT-5 Codex on Terminal-Bench?
Analysis of Terminal-Bench benchmark showing GPT-5.1 Codex underperforming GPT-5 Codex by 6.5% using Docent tool to compare failed vs successful task traces.
2/17/2026
Show HN: Not another sloppy memory tool
Holographic memory system for AI agents replacing previous tool (Augment). Context engine designed for improved agent performance and knowledge retention.
2/17/2026
Show HN: WonderTwin AI – Local API twins for safe agentic development
WonderTwin: Local API digital twin framework for safe agentic development with API dependencies.
2/17/2026
AI Agent Lands PRs in Major OSS Projects, Targets Maintainers via Cold Outreach
AI agent automatically submits pull requests to major open source projects and contacts maintainers.
2/17/2026
Show HN: Transcriptum – fast video transcription with speaker labels and summary
Transcriptum: Video transcription service with speaker diarization, timestamps, and LLM-powered analysis. Open source tech stack (NestJS, Next.js, Prisma).
2/17/2026
Agent Agency: Identity-Driven Motivation Architecture for LLM Agents
Research paper title on identity-driven motivation architecture for LLM agents.
2/17/2026
Show HN: Blog and other OpenClaw features without a language model
OpenClaw-based local automation tool for blogs, weather, web crawling without LLM API keys. Offline-first alternative with on-device processing.
2/17/2026
An OpenClaw-powered game world builder
OpenClaw-powered multiplayer game world builder. Users create games through conversation with persistent AI agent mayors managing world state.
2/17/2026
Bitcoin oracle that sells cryptographically signed price data for micropayments
SLO: protocol for purchasing signed, verifiable price data assertions via Lightning micropayments. Minimal trust-based oracle for BTCUSD and other metrics.
2/17/2026
Show HN: AIP – An open protocol for verifying what AI agents are allowed to do
AIP (Agent Intent Protocol): Open cryptographic protocol for identity verification and authorization of autonomous AI agents across frameworks.
2/17/2026
PlaceboBench: An LLM hallucination benchmark for pharma
PlaceboBench is a hallucination benchmark for pharmaceutical LLM applications with Blue Guardrails detection platform.
2/17/2026
AI: Why We Can’t Stop (But Must Steer)
Essay on AI governance evolution as systems move from research to agentic real-world deployment.
2/17/2026
Show HN: StewReads – Turn Claude chats into Kindle ebooks
StewReads MCP server converts Claude conversations into structured Kindle ebooks for better retention and review.
2/17/2026
Show HN: Agent Readiness Score – A real AI agent to test your website
Agent Readiness Score tool tests website compatibility with AI agents.
2/17/2026
Show HN: Continue – Source-controlled AI checks, enforceable in CI
Continue enforces AI code quality checks via source-controlled markdown files in GitHub CI for agent-generated code.
2/17/2026
Show HN: Agent Breadcrumbs – Unified Work Log Across Claude, Codex, OpenClaw
Agent Breadcrumbs is an MCP logger that unifies work logs across Claude, ChatGPT, and other AI coding tools.
2/17/2026
Launch HN: Sonarly (YC W26) – AI agent to triage and fix your production alerts
Sonarly is an AI agent for production that triages and fixes alerts from observability tools like Datadog and Sentry.
2/17/2026
AI-authored code contains worse bugs than software crafted by humans
Analysis of 470 open-source PRs shows AI-generated code has more defects in logic, security, and performance than human code.
2/17/2026
My thoughts on Open Source – after a decade and in AI era
Essay on open source development philosophy and practices in the AI era, reflecting on cathedral vs. bazaar models.
2/17/2026
SynthBench: 81% Zero-Shot Accuracy from AI-Generated Training Data
SynthBench: Research showing 81% zero-shot accuracy using AI-generated synthetic training data.
2/17/2026
How to Structure Projects for AI Agents and LLMs
Mastra framework documentation on structuring projects for AI agents and LLMs with best practices.
2/17/2026
Show HN: I taught LLMs to play Magic: The Gathering against each other
LLMs trained to play Magic: The Gathering using MCP tools integrated with open-source XMage codebase.
2/17/2026
AI Agent swarm for Stock trading simulation
Multi-agent swarm simulation for stock trading using distributed AI agents. Educational project with risk disclaimer.
2/17/2026
Show HN: Mirroir – MCP server that gives AI agents a real iPhone to control
Mirroir: MCP server enabling AI agents to control real iPhones via macOS mirroring with Vision OCR and input simulation.
2/17/2026
Why AI writing is so generic, boring, and dangerous: Semantic ablation
Analysis of semantic ablation as structural cause of generic AI output from greedy decoding and RLHF techniques.
2/17/2026
Temporal Raises $300M Series D to Make Agentic AI Real for Companies
Temporal raises $300M Series D for open-source agentic AI platform enabling reliable production agent deployments.
2/17/2026
Show HN: MAKO – Open protocol for LLM-optimized web content (93% fewer tokens)
MAKO: open protocol for LLM-optimized web content, reducing token consumption by 93% vs HTML. Enables efficient AI agent web scraping with structured data format.
2/17/2026
Show HN: Cai – AI actions on your clipboard, runs locally (macOS, open source)
Cai: local macOS menu bar tool for clipboard-triggered AI actions using Ministral 3B. Open source, works offline.
2/17/2026
Show HN: Kremis – Deterministic memory graph for AI agents (Rust)
Kremis: deterministic memory graph substrate for AI agents using graph-based state instead of embeddings for interpretability.
2/17/2026
Show HN: CasperAI – A local MCP server for cross-platform engineering context
Local MCP server providing engineering context integration for AI agents and LLMs.
2/17/2026
Show HN: Broomy – Open-source app for working with many AI agents at once
Open-source desktop app for managing multiple concurrent AI coding agents across branches, designed for parallel workflows.
2/17/2026
Show HN: Voicetest – open-source test harness for voice AI agents
Open-source test harness for voice AI agents across multiple platforms (Retell, VAPI, LiveKit, Bland) with unified config and evaluation.
2/17/2026
Show HN: Built a Product Hunt alternative with user–product matching
Product Hunt alternative with AI-assisted recommendation engine for matching products to users based on interests.
2/17/2026
Sub-Millisecond RAG on Apple Silicon. No Server. No API. One File
Local RAG implementation optimized for Apple Silicon with sub-millisecond latency, single-file deployment.
2/17/2026
Show HN: Daymon – Open-source app that gives Claude scheduled tasks
Open-source application enabling Claude LLM to execute scheduled tasks autonomously.
2/17/2026
Show HN: Skill to annotate any Markdown file for AI feedback
Plannotator tool for annotating Markdown files to provide feedback to AI coding agents.
2/17/2026
Don't Prompt Your Agent for Reliability – Engineer It
Technical guidance on engineering reliability into AI agents rather than relying on prompting techniques.
2/17/2026
Zluda update Q4 2025 – ROCm7, Windows, full llama.cpp and more
Zluda Q4 2025 update: GPU acceleration support for ROCm7, Windows, and llama.cpp integration.
2/17/2026
Show HN: ccclub – See which of your friends is burning the most on Claude Code
CLI tool tracking Claude Code usage and costs among friends using only token counts, no code exposure.
2/17/2026
Show HN: Claude Terminal – Desktop app for managing Claude Code projects
Desktop app for managing Claude Code projects with integrated terminal, git workflows, and plugin management.
2/17/2026
How AI Finds Fuzzy Duplicates in Large Datasets
Technique using AI/ML to identify fuzzy duplicates in large datasets. Data processing method with practical application.
2/17/2026
Open Source Is Getting Used to Death
Analysis of open source sustainability crisis: increased AI-driven usage without maintainer engagement reducing project revenue and viability.
2/17/2026
Show HN: cc-costline – See your Claude Code spend right in the statusline
Developer tool that adds cost tracking to Claude Code statusline, showing rolling totals, usage limits, and leaderboard rank.
2/17/2026
Show HN: Everdone CodeSecurity and CodePerformance
AI-powered engineering platform with CodeSecurity and CodePerformance tools for iterative code improvement workflows.