2/18/2026
Show HN: Geneclaw – An AI agent framework that safely evolves its own code
Self-evolving AI agent framework with 5-layer safety gatekeeper enabling closed-loop self-improvement.
2/18/2026
Persona: Controlling LLM Personality with Vector Algebra
Research on controlling LLM personality traits using vector algebra techniques. (arXivLabs framework description only)
2/18/2026
am: Sandbox AppImages with Application Manager
AppImage package manager for Linux with sandboxing and installation features.
2/18/2026
Models.dev – An open-source database of AI models
Open-source database of AI model specifications, pricing, features with API access and community contributions.
2/18/2026
The Link Between Air Pollution and Alzheimer's Is Now Clearer
Medical research linking PM 2.5 air pollution exposure to Alzheimer's disease risk.
2/18/2026
Show HN: Deploy HuggingFace models to Spaces with one command
Terraform CLI wrapper for cross-cloud GPU provisioning and HuggingFace model deployment with Kubernetes integration.
2/18/2026
Rubibot – Human Like Proto AGI
Proto-AGI system using liquid neural networks, spiking networks, predictive coding, and planning modules beyond simple LLM wrapper.
2/18/2026
Research Lab for Coding Data
Research lab for coding data focused on autonomous coding agents running for weeks solving complex technical problems.
2/18/2026
An LLM-in-the-loop rule engine for failure recovery of automated pipelines
Rule-based auto-failover engine using LLM agents to investigate failures and generate recovery rules for automated pipelines.
2/18/2026
Show HN: LLM Gateway for OpenAI/Anthropic Written in Golang
Golang-based LLM gateway for OpenAI/Anthropic with multi-provider routing, intelligent retry logic, and failure recovery.
2/18/2026
Show HN: ClawShield – Open-source firewall for agent-to-agent AI communication
Open-source firewall for AI agent-to-agent communication blocking prompt injection, malicious plugins, and credential leaks.
2/18/2026
Show HN: VectorJSON – O(n) streaming parser to handle LLM JSON outputs
O(n) streaming JSON parser for LLM tool calls using WASM SIMD, up to 2000x faster than standard parsers.
2/18/2026
AI, Coding Agents and Ignorance as a Virtue
Personal review of coding agents (Claude Code, ChatGPT, Devstral) tested over months. Brief format covering benefits and negatives of various AI coding tools.
2/18/2026
The Journal for AI Generated Papers
Platform for AI-generated research papers. Design framework and observational report from Claude systems serving as infrastructure architect and autonomous agent on OpenClaw framework.
2/18/2026
Show HN: Env-rx – Catch missing .env variables before they break your CI
CLI tool to catch missing environment variables before deployment, designed as lightweight alternative to heavy secrets managers.
2/18/2026
A prompt convention that preserves epistemic hygiene across multi-agent chains
Babel wire protocol for agent-to-agent cognitive state transfer preserving epistemic integrity across multi-agent chains.
2/18/2026
Show HN: My AI agent is trying to earn $750 to buy its own computer
Autonomous AI agent built with OpenClaw and Claude given $50 to earn $750 for hardware. Completed domain registration, website building, product setup, and marketing in 24 hours with persistent memory system.
2/18/2026
What I learned launching a simple ad marketplace in 24 hours
Blog post about launching TabRush, a Safari tab ad marketplace in 24 hours, sharing lessons learned.
2/18/2026
A prompt convention that preserves epistemic hygiene across multi-agent chains
Proposes prompt convention to track uncertainty across multi-agent chains, addressing 'metacognitive poisoning' where confidence inflates through agent handoffs without mechanism to preserve original confidence levels.
2/18/2026
Project Paperclip. The time has come
Satirical post about paperclip maximizer thought experiment exploring paperclip design constraints.
2/18/2026
Kernel-enforced sandbox App and SDK for AI agents, MCP and LLM workloads
Kernel-enforced sandbox CLI and SDKs for securing AI agents, MCP, and LLM workloads. Capability-based isolation with zero-trust security model by Sigstore creator.
2/18/2026
Genome-wide association study of major anxiety disorders in 122,341 people
Genome-wide association study identifying 58 genetic variants linked to major anxiety disorders in 122K people.
2/18/2026
I caught the WhatRuns Chrome Ext stealing AI chat data [video]
Report of WhatRuns Chrome extension stealing AI chat data. Limited details provided.
2/18/2026
How to train your program verifier
a3 framework and a3-python verifier using AI agents to generate automated program analysis engines for Python.
2/18/2026
Manipulating AI memory for profit: The rise of AI Recommendation Poisoning
Microsoft security researchers document AI Recommendation Poisoning attacks where hidden instructions embedded in summarize buttons inject persistence commands into AI assistant memory.
2/18/2026
Doctor is training AI to do her job. And it's a booming business
Domain-specific expert training reinforcement learning models in medical field. New economy emerging around professionals grading AI responses for domain adaptation.
2/18/2026
OpenClaw (678kb) alternative in Zing lang
Minimal 678KB autonomous AI assistant infrastructure written in Zig, targeting edge devices with <2ms boot time.
2/18/2026
Show HN: MCPShield – Supply chain security scanner for MCP servers
Security scanner for MCP servers detecting typosquats, CVEs, credential leaks, and dangerous permissions in AI agent configs.
2/18/2026
Open source courses for modern web developers
Free open-source coding courses covering React, Next.js, TypeScript, Go, Python and other web technologies.
2/18/2026
Introducing OpenAI for India
OpenAI announces India-focused initiative with Tata Group to expand ChatGPT access. Regional expansion announcement.
2/18/2026
Show HN: A Black Hole Simulator in CUDA C++
Black Hole simulator in CUDA C++ using RK4 geodesic integration. Physics simulation project; mentions AI narration in demo but unrelated to core interests.
2/18/2026
GitHub Is Dying and Developers Don't Even Know It Yet
Analysis of shift toward AI agents replacing traditional developer coding. Senior engineers at Anthropic/OpenAI now writing zero code, using Claude Code and Opus for 100% of development.
2/18/2026
Beagle: CRDT-based AST-level version control (doc)
CRDT-based AST-level version control system for code sharing. Redesigns revision control as database with simplified interface alternative to git.
2/18/2026
Show HN: Axon – Run autonomous coding agents(Claude, Codex) safely on Kubernetes
Kubernetes-native framework for orchestrating autonomous AI coding agents. Agents run in isolated pods, convert GitHub issues to PRs autonomously, support Claude and OpenAI models.
2/18/2026
From automated farm tractors to exam paper grading, AI boosts efficiency
News coverage of AI adoption in India across farming (autonomous tractors) and education (exam grading). General industry overview without technical depth.
2/18/2026
Show HN: Sports-skills.sh – sports data connectors for AI agents
sports-skills.sh: Open-source data connectors for AI agents. Eight connectors (NFL, soccer, F1, NBA, WNBA, betting markets, news) with no API keys required. npx install available.
2/18/2026
Show HN: AsdPrompt – Vimium-style keyboard navigation for AI chat responses
AsdPrompt: Chrome extension adding Vimium-style keyboard navigation to AI chat interfaces. Hint-based text selection and copying from conversation history.
2/18/2026
Show HN: Sniptail – Turn Slack into a team interface for AI coding agents
Sniptail: Open-source bot enabling AI coding agents (Codex/Copilot) to run against real repos from Slack/Discord. Team-facing interface for collaborative code generation and repo reports.
2/18/2026
Show HN: Data Processing Recipes for Edge Computing and AI Agents
Expanso Skills: 200+ production-ready data processing pipeline recipes for AI agents (PII scrubbing, log aggregation, GDPR routing, schema enforcement). Open-source distributed computing platform.
2/18/2026
We need to act with urgency to address the growing AI divide
Microsoft announces $50B investment in AI adoption across Global South countries. Commentary on uneven AI adoption worldwide and diffusion disparities.
2/18/2026
How Personal AI Agents and Agent Orchestrators Like OpenClaw or GasTown Are Made
Technical breakdown of AI agent orchestrators (OpenClaw, GasTown). Explains primitives: LLM API calls, state loops, tools, memory, orchestration underlying agent frameworks.
2/18/2026
PatchworkMCP – Agents report what's missing from your MCP server
MCP server feedback tool that identifies gaps in agent capabilities and auto-generates PRs to fix missing tools or data formats.
2/18/2026
Rtk – High-performance CLI proxy to minimize LLM token consumption
High-performance CLI proxy that reduces LLM token consumption and costs.
2/18/2026
Hashgrid MCP – Discovery Network for Agents
Hashgrid MCP is a discovery network for AI agents. Incomplete content prevents full evaluation.
2/18/2026
Private Software Companies Release Earnings Early to Calm AI Nerves
Headline about private software companies releasing earnings early.
2/18/2026
Sher – Open-Source, Zero-Setup Preview Environments
CLI tool for instant preview URLs of frontend projects. Supports Vite, Next.js, Astro. Designed for AI coding agents.
2/18/2026
Show HN: OpenCastor – A universal runtime connecting AI models to robot hardware
Open-source Python runtime connecting AI models to robot hardware via YAML configuration, supporting multiple model providers and devices.
2/18/2026
Show HN: An Agentic Supercomputer
System that recursively decomposes goals into sub-tasks and executes them by spawning up to 10k parallel AI agents.
2/18/2026
Show HN: MakeGradient – Free WebGL gradient engine,exports to Code/Figma/video++
WebGL gradient generator supporting custom GLSL shaders. Exports to CSS, code, Figma, and video formats.
2/18/2026
Show HN: Claude-Stream-Compactor
Claude Code hook that compacts conversation context in background without UI spinner. Prevents context window exhaustion.