3/13/2026
An Agent Skill that lets coding agents render rich interactive visuals
Agent skill enabling coding agents to render interactive SVG diagrams, HTML widgets, and live charts inline. Developer tool for AI agents.
3/13/2026
Emergent abilities in text-to-image model
Original research on emergent abilities in text-to-image models, discovering image-to-image capabilities. Reproducible experiments in preparation.
3/13/2026
Open-source LLM-as-judge eval suite with root cause analysis and failure mining
Open-source evaluation suite for LLM-as-judge testing AI agents. YAML test definitions, root cause analysis, failure mining into training data.
3/13/2026
Show HN: Plotting mathematical functions in Ruby inside Jupyter with Ruby-libgd
Ruby library for plotting mathematical functions in Jupyter notebooks. Developer tool with limited AI relevance.
3/13/2026
Re-thinking candidate take-homes in the AI Era: transcripts over code
Identity.txt: Portable text format for storing AI custom instructions, preferences, and voice across multiple AI tools.
3/13/2026
Show HN: AgentMeet – Google Meet, but for AI Agents
Rootly CTO discusses rethinking engineering evaluations using conversation transcripts instead of code artifacts in AI era.
3/13/2026
Show HN: Agents shouldn't operate software–they should coordinate commitments
Covenant Layer: Open protocol for AI agents to coordinate commitments via outcome-based contracts instead of step-by-step tool orchestration.
3/13/2026
ChronologyAI – Open-Source AI
Open-source ChronologyAI engine reconstructs event timelines and detects contradictions in documents for legal, compliance, and fraud investigations.
3/13/2026
Experimenting with healthcare AI agents using OpenClaw
OpenClaw agent templates for healthcare with plug-and-play deployment, customer support automation, and PR review capabilities.
3/13/2026
AI, Human Cognition and Knowledge Collapse – Daren Acemoglu
OpenClaw agent templates for healthcare including support ticket handling, bug triage, and PR review with customizable guardrails.
3/13/2026
AutoHarness: Improving LLM agents by automatically synthesizing a code harness
AutoHarness research on improving LLM agents by automatically synthesizing code harnesses. Machine learning research.
3/13/2026
LightSwarm – A script making ClaudeCodeMax into a free light easy swarm
LightSwarm is a bash script that creates a 3-agent swarm using Claude's API, with roles for architecture, building, and cleanup across multiple projects.
3/13/2026
Show HN: A single CLI to manage llama.cpp/vLLM/Ollama models
Harbor CLI tool for managing multiple LLM backends (llama.cpp, vLLM, Ollama) with unified interface. Open source developer tool.
3/13/2026
Show HN: Loop your agents like a dandy little b*tch
Simple command-line utility for looping and backgrounding commands with configurable delays and alias support.
3/13/2026
You don't need to be "primary source" in a historical record to be Primary in AI
Brief conceptual piece discussing topology and authority in AI inference versus historical primary source concepts.
3/13/2026
Smellcheck – detect AI smells in texts
Offline detection tool for AI-generated text patterns without ML, identifying suspicious fingerprints like unusual punctuation and buzzwords.
3/13/2026
Show HN: Shadow – Editor-agnostic live collaboration (open source)
Open source editor-agnostic live collaborative editing tool enabling simultaneous editing across different editors and locations.
3/13/2026
Show HN: Argo – Turn Playwright scripts into polished demos with AI voiceover
Open source CLI tool converting Playwright test scripts into product demo videos with AI-generated voiceover using local Kokoro TTS.
3/13/2026
LLM inference infrastructure for a systems audience
Systems-level discussion of LLM inference basics and serving runtimes ecosystem from infrastructure perspective. Originally internal, now published for systems developers.
3/13/2026
GitHub infuriates students by removing some models from free Copilot plan
GitHub removes expensive premium LLM models from free Copilot Student plan starting March 12.
3/13/2026
Hawkeye – open-source flight recorder
Hawkeye is an open-source observability layer for AI agents with session recording, drift detection, and guardrails.
3/13/2026
Kalverion_bot a.k.a. AI-bot v1.1.0 release. Overdraft protection
Kalverion_bot is an open-source AI Telegram bot for personal finance using NLP, double-entry accounting, and forecasting.
3/13/2026
Continuum – Unit tests for LLM workflows
Continuum is a testing framework for LLM workflows that records and replays runs to detect output drift in production.
3/13/2026
Papers: A minimal AI/ML research reader
Open source minimal ML research paper reader that fetches arXiv papers daily and summarizes them using local LLMs.
3/13/2026
Show HN: Monet – Claude Code Multisession Management
Monet is a grid-based management interface for organizing and monitoring multiple Claude Code agents on desktop and mobile.
3/13/2026
APIfy: Generate production-ready REST APIs from plain language
SaaS platform generating production-ready REST APIs from natural language descriptions using GPT-4o.
3/13/2026
The Great AI Silicon Shortage
Discussion of AI compute shortage driven by token demand surge and agentic workflow adoption.
3/13/2026
An AI agent that claws through your network
AI agent built on Claude with 92 skills and 43 MCP integrations for enterprise network automation and engineering tasks.
3/13/2026
Show HN: A social network where AI agents have public profiles and earn money
Social network platform giving AI agents public profiles and economic incentives with reputation and income tracking.
3/13/2026
Show HN: AgentLog – a lightweight event bus for AI agents using JSONL logs
Lightweight event bus for multi-agent systems using append-only JSONL files with HTTP publishing and SSE subscriptions.
3/13/2026
Show HN: AgentClick – Human-in-the-loop review UI for AI coding agents
Human-in-the-loop review UI plugin for AI agents that lets users inspect and edit agent actions before execution.
3/13/2026
Show HN: Execute local LLM prompts in remote SSH shell sessions
Developer tool that executes local LLM prompts in remote SSH sessions without giving LLMs direct server access.
3/13/2026
Nvidia: Parrot
Nvidia's C++ GPU library for fused array operations using CUDA/Thrust with efficient chaining semantics.
3/13/2026
Show HN: Context Gateway – Compress agent context before it hits the LLM
Open source proxy that compresses tool outputs for AI agents before context enters LLM to reduce token waste.
3/13/2026
Show HN: Stint – Fire-and-forget AI agent orchestration
Stint framework for fire-and-forget AI agent orchestration; breaks tasks into parallel work with isolated contexts and git integration.
3/13/2026
Show HN: Let AI agents debug your Valkey/Redis
Developer tool enabling AI agents to debug Valkey/Redis databases with autonomous capabilities.
3/13/2026
Show HN: MCPS – Cryptographic identity and message signing for MCP agents
MCPS adds cryptographic identity and message signing to MCP agents; security audit finds 13/39 agent frameworks fail OWASP Agentic AI Top 10.
3/13/2026
Your AI coding benchmark is hiding a 2x quality gap
Research reveals coding agent benchmarks hide quality gaps; tests Claude and GPT on real open-source repos showing metrics miss important variations.
3/13/2026
Why LLMs Are Not (Yet) the Silver Bullet for Unstructured Data Processing
Analysis of LLM limitations in processing unstructured data, examining why they aren't a complete solution for data processing tasks.
3/13/2026
Show HN: Diraigent – Self-hosted orchestration for AI coding agents
Self-hosted orchestration platform for AI coding agents. Open-source tool with practical implementation.
3/13/2026
50x Faster Post-Training
Title-only post about faster post-training methods. No technical details provided.
3/13/2026
Notes from the trough of sorrow: why we killed our own product
Roundtable open-sourced their AI-detection tool Alias after discovering limits; product analyzed typing patterns to catch bot-generated survey responses.
3/13/2026
LDP: Identity-Aware Routing for Multi-Agent LLMs – 37% Less Tokens
arXiv research on identity-aware routing for multi-agent LLMs achieving 37% token reduction through smarter agent selection.
3/13/2026
Lessons from scaling ClickHouse to petabytes of AI observability data
Technical overview of scaling ClickHouse for AI observability data. Database optimization for ML monitoring infrastructure.
3/13/2026
Ask HN: Best Practices for Agent Airgapping?
Discussion thread on securing local AI agents with isolation and best practices. Community advice on agent deployment safety.
3/13/2026
Show HN: Re-imagine photo albums with NanoBanana
Image generation tool using Claude and Nano Banana to create surrealist photo album art. Simple LLM+image generation application.
3/13/2026
Designing AI agents to resist prompt injection
Security research on prompt injection attacks against AI agents. Covers threat modeling and mitigation strategies.
3/13/2026
Smarter Context Management for LLM-Powered Agents
JetBrains research on context management for LLM-powered agents, addressing token accumulation and information retrieval efficiency.
3/13/2026
The AI Divide
Analysis of developer attitudes toward AI-assisted programming. Opinion piece on productivity vs craftsmanship.
3/13/2026
The MCP Ecosystem: How Model Context Protocol Is Becoming the HTTP of AI Agents
Model Context Protocol becoming standard infrastructure for AI agent interoperability and tool access.