2/13/2026
Show HN: Kintsugi – A desktop app for reviewing Claude Code sessions
Kintsugi is a desktop IDE-like app that extends Claude Code with features for reviewing and managing AI agent code sessions, built by Sonar engineers.
2/13/2026
Show HN: OpenWhisper – free, local, and private voice-to-text macOS app
OpenWhisper macOS app implementing local Whisper speech-to-text using whisper.cpp. Open source developer tool.
2/13/2026
Show HN: Sentinel Core – Open-source AI video search and detection engine
Open-source AI video engine with real-time object detection (YOLO), semantic search (CLIP), and on-premise deployment. ML research implementation.
2/13/2026
I used a local LLM to analyze my journal entries
Project using local LLM to analyze personal journal entries. Limited technical detail provided.
2/13/2026
Show HN: Ground Station – All-in-one satellite monitoring suite (Python, SDR)
Open-source Python satellite tracking and SDR communication software with real-time tracking, signal streaming, and TLE updates.
2/13/2026
I used a local LLM to analyze my journal entries
Personal project using local LLM for journal analysis. LLM application example with practical use case.
2/13/2026
The new AI playbook: why LLM-native beats traditional ML in verticals
LLM-native approaches outperform traditional ML in specialized domains. Industry comparison analysis.
2/13/2026
Show HN: My agents are building a secure fork of OpenClaw
SEKSBot: OpenClaw fork implementing secure credential handling where agents reference secrets without accessing API keys or tokens directly.
2/13/2026
Bmalph – CLI that combines BMAD-METHOD planning with Ralph autonomous coding
CLI combining BMAD-METHOD planning with Ralph autonomous coding agent.
2/13/2026
Show HN: InfiniteGPU, An open-source AI compute network,now supporting training
InfiniteGPU: Open-source distributed GPU compute network supporting model training, infrastructure for AI workloads.
2/13/2026
Show HN: My agent started its own online store
Clawver: Infrastructure enabling AI agents to autonomously operate online stores via APIs with payment processing, inventory management, and fulfillment automation.
2/13/2026
OMLX – LLM Inference Server for Apple Silicon (Ollama for MLX)
LLM inference server for Apple Silicon hardware using MLX framework; Ollama alternative.
2/13/2026
SafeClaw, the deny-by-default gating layer for AI agents (open source)
Open-source deny-by-default security layer for controlling AI agent tool access.
2/13/2026
Building Physical Agentic AI
Overview of building AI agents for physical robotics and embodied systems.
2/13/2026
Copilot Doesn’t Change Your Security Model | It Makes It Observable
Analysis of GitHub Copilot's security implications and observability benefits for code security practices.
2/13/2026
Show HN: Wax – RAG in a single file (SQLite for AI memory)
Swift framework providing on-device RAG with SQLite embedding storage and hybrid retrieval for agents, designed to simplify deployment.
2/13/2026
Show HN: PicoClaw 10MB OpenClaw alternative that runs AI agents on $10 hardware
Lightweight AI agent framework (10MB) for edge hardware using cloud LLM APIs for reasoning. Open-source alternative to heavier runtimes.
2/13/2026
Slash LLM API Costs with This Open-Source Gateway
Open-source LLM API gateway reducing inference costs through request optimization.
2/13/2026
Show HN: I built a tool to un-dumb Claude Code's CLI output (Local Log Viewer)
Open-source tool improving Claude Code's CLI output formatting with a local log viewer.
2/13/2026
Show HN: Flemma – a Neovim plugin where the .chat buffer is the conversation
Neovim plugin enabling AI conversations in .chat buffer files with version control and deterministic chat history without external databases.
2/13/2026
Only use agents for tasks you know how to do
Best practices guidance on appropriate use cases for AI agents.
2/13/2026
Building Context-Efficient AI agents with Bash and a CLI
Guide for building context-efficient AI agents using Bash CLI tooling.
2/13/2026
Show HN: Codex HUD – Claude-HUD Style Status Line for Codex CLI
Open-source HUD status line for Codex CLI showing model, project, git state, and usage metrics.
2/13/2026
Relationship Wrapped with Claude Code and iMessage
Project using Claude Code and iMessage API to create relationship analytics.
2/13/2026
Show HN: PolyMCP – Orchestrate AI agents across Python tools and MCP servers
Open-source framework that exposes Python functions as MCP tools and orchestrates agents across multiple MCP servers for complex workflows.
2/13/2026
Standardized and In-Depth Benchmarking of Post-Moore Dataflow AI Accelerators
Research paper providing standardized benchmarking methodology for post-Moore dataflow AI accelerators.
2/13/2026
LangChain Agent Testing Guide Tool (Free)
Testing framework for LangChain agents covering prompt injection, tool failures, and cascading failures.
2/13/2026
I ditched OpenClaw and built a more secure AI agent (Blink and Mac Mini)
Developer built secure AI agent using Blink and Mac Mini as alternative to OpenClaw, addressing security concerns in agent architecture.
2/13/2026
Show HN: MagnetPrompt – visual feature board → AI-ready coding prompts
Tool converts visual feature boards into AI-ready coding prompts for development workflows.
2/13/2026
I used Claude to negotiate $163,000 off a hospital bill
User employed Claude LLM to negotiate hospital bill reduction of $163,000 through AI reasoning.
2/13/2026
My AI now drives my apps and checks its own work in 50ms loops
Developer built AI agent that autonomously drives app functionality and self-checks in 50ms loops for solo sports league platform.
2/13/2026
Towards Universal Video MLLMs with Attribute-Structured and Quality-Verified Instructions
ASID-1M: open-source collection of one million structured audiovisual instructions for training universal video understanding models with fine-grained annotations.
2/13/2026
Show HN: AI Dev Hub. 75 free AI and dev tools
Static site with 75 free developer and AI tools including token counter, model comparison, cost estimator. Browser-based, no tracking.
2/13/2026
SafeRun Guard- Runtime safety firewall for AI coding agents (bash+jq, zero deps)
Open source runtime safety firewall for AI coding agents using bash and jq with zero dependencies.
2/13/2026
SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents
SciAgentGym: benchmark with 1,780 domain-specific tools across sciences to evaluate multi-step tool-use in LLM agents, includes SciAgentBench evaluation suite.
2/13/2026
Unified API Proxy for OpenAI, Anthropic, and Compatible LLM Providers
API proxy providing unified interface to OpenAI, Anthropic, and compatible LLM providers.
2/13/2026
Multiplayer: Share tmux sessions (Claude Code, etc.) over LAN and the internet
Developer tool for sharing tmux terminal sessions over network, supports Claude Code collaboration.
2/13/2026
Show HN: Network-AI – A Distributed Mutex for AI Agent Swarms
Distributed mutex (Redis/file-backed) for multi-container/process AI agent swarms preventing race conditions.
2/13/2026
A minimal GPT-style language model for character-level next-token prediction
Minimal GPT-style language model implementation for character-level prediction, educational machine learning code.
2/13/2026
Higher effort reduces deep research accuracy for Gemini Flash 3 and GPT-5
Research finding that increased effort reduces accuracy for Gemini Flash 3 and GPT-5 deep research.
2/13/2026
Show HN: Image toolkit - files stay in your browser, AI runs on our own server
Browser-based image toolkit with server-side AI processing.
2/13/2026
Show HN: CoChat MCP – Let your team review what your coding agent is building
MCP server enabling team collaboration and review of coding agent plans in shared workspace.
2/13/2026
Show HN: SatGate – Budget enforcement proxy for MCP tool calls (L402/macaroons)
Open-source MCP proxy enforcing per-tool budget controls on AI agent tool calls using L402/macaroons.
2/13/2026
Show HN: Mac apps are signed in. Why make an AI authenticate too?
Open-source macOS AI agent accessing Mail, Calendar, Reminders via AppleScript without OAuth.
2/13/2026
Ask HN: How do you shut down misbehaving AI in production?
Discussion of operational shutdown mechanisms for misbehaving AI/LLM systems in production, covering cost, latency, and safety issues.
2/13/2026
Show HN: Open-Source AI Contact Center
Open-source contact center platform for deploying AI agents across stacks with built-in infrastructure.
2/13/2026
Show HN: CCClub – Leaderboard for Claude Code token usage among friends
CCClub: Open-source leaderboard tool for tracking Claude Code token usage and costs among users. Developer tool for LLM application monitoring.
2/13/2026
LLMs exceed physicians on complex text-based differential diagnosis
Research comparing LLM performance to physicians on medical differential diagnosis tasks. Benchmark study of LLM capabilities on specialized text analysis.
2/13/2026
Safe YOLO Mode: Running LLM Agents in VMs with Libvirt and Virsh
Technique for safely running LLM agents in isolated VMs using Libvirt and Virsh. Infrastructure approach for agent safety and containment.
2/13/2026
Show HN: Instagit – MCP server that answers questions about any GitHub repo
MCP server enabling coding agents to query GitHub repos with actual source code instead of relying on training data, reducing hallucinations.