3/15/2026
LLM FFN benchmarks on a 4‑core HP All‑in‑One
rolvsparse is a compute primitive for matrix arithmetic in AI processors, claiming up to 133.5× speedup on Llama models with 99.9% energy reduction on existing hardware.
3/15/2026
Nova–Self-hosted personal AI that learns from corrections and fine-tunes itself
Nova is an open-source personal AI that runs locally, learns from user corrections, and fine-tunes itself without cloud dependencies or data leaving the device.
3/15/2026
Show HN: Open-source playground to red-team AI agents with exploits published
Open-source playground for red-teaming and testing AI agents with published exploits to improve agent robustness and trustworthiness.
3/15/2026
Token optimized query language (Memelang)
Memelang is an AI-optimized query language that reduces token count and model size for LLM RAG applications; free for development use.
3/15/2026
Show HN: Scitex-notification – Give AI agents a voice: TTS, phone calls, SMS
scitex-notification adds multi-channel notifications (TTS, phone calls, SMS) to AI agents via escalating alerts when tasks complete.
3/15/2026
Show HN: Kubernetes Security Profile Generator Using eBPF
kguardian uses eBPF to automatically generate Kubernetes security profiles and NetworkPolicies by monitoring actual container behavior.
3/15/2026
CostClaw – free local dashboard to track and reduce OpenClaw agent costs
CostClaw is a free open-source plugin for OpenClaw agents that tracks LLM API costs in real-time via a local dashboard without external services.
3/15/2026
Copilot Health
Copilot Health helps people understand medical test results and health data through AI-powered interpretation and consultation support.
3/15/2026
Multi-agent fleet management for coding agents
Open-source AgentFactory orchestrates fleet of coding agents (Claude, Codex, Spring AI) through automated pipeline for issue resolution and code shipping.
3/15/2026
OpenJarvis: Personal AI, on Personal Devices
Open-source framework for personal AI agents running entirely on-device with efficiency-aware evaluations and learning loop using local trace data.
3/15/2026
Show HN: Free OpenAI API Access with ChatGPT Account
NPM package enabling free OpenAI API access via ChatGPT OAuth tokens. Creates localhost proxy to ChatGPT backend API with Vercel AI SDK provider support.
3/15/2026
I'm Too Lazy to Check Datadog Every Morning, So I Made AI Do It
AI automation tool to summarize Datadog monitoring alerts and escalate issues, reducing manual dashboard review.
3/15/2026
LLMs can be absolutely exhausting
Discussion of mental fatigue and workflow challenges when working with LLMs like Claude and Codex, and recovery strategies.
3/15/2026
Claude-Code-Workflow – Orchestrate Multiple CLI Agents
Multi-agent workflow orchestration system supporting Gemini, Qwen, Claude with role-based agents, background execution, and visual workflow editing.
3/15/2026
Continuum – GitHub Action that detects LLM drift in CI
GitHub Action detecting LLM output drift in CI/CD by replaying workflows and diffing outputs to prevent silent model changes reaching production.
3/15/2026
Show HN: ObservAgent – Observability for Claude Code(cost, tools, subagents)
Dashboard for real-time observability into Claude Code sessions, tracking costs, tool usage, and subagent execution without code changes.
3/15/2026
Faith Claw – Security middleware for autonomous AI agents (OpenClaw)
Security middleware for autonomous AI agents that risk-scores actions, detects injection attacks, and catches behavioral drift across multi-turn interactions.
3/15/2026
Kuberna Labs – Open-source SDK for autonomous cross-chain AI agents
Open-source SDK for building autonomous AI agents that execute cross-chain financial operations with cryptographic guarantees and trusted execution environments.
3/15/2026
Multi-agent coordination via timer-based Discord polling (Claude Code)
Multi-agent coordination system using Claude Code, Discord webhooks, and timer-based polling. Production autonomous workflows with real-time notifications.
3/15/2026
Show HN: Claude's 2x usage promotion (March 2026) in your timezone
Timezone converter tool for Claude API usage promotion (Mar 2026). Minor LLM-adjacent utility.
3/15/2026
Securing AI Agents
Overview of layered security architecture for AI agents, emphasizing secure human identity verification and token-based authorization.
3/15/2026
Show HN: Quell, a local security layer to stop AI IDEs leaking your secrets
Quell is a local security layer that intercepts prompts to AI IDEs, redacting secrets before they reach cloud models, storing values in OS keychain.
3/15/2026
Show HN: ARISE – Agents that create their own tools at runtime when they fail
ARISE framework enables LLM agents to synthesize their own tools at runtime when they encounter task gaps, adapting without pre-crafted tool libraries.
3/15/2026
Show HN: Turn any file into a CLI (reduce tokens vs. MCP)
clifast tool converts TypeScript/JavaScript functions into CLI packages with optimized help text for LLM navigation, reducing token usage versus MCP.
3/15/2026
I Built LiveAuth: POW and Lightning Network Authentication for AI Agents
LiveAuth system providing Proof-of-Work and Lightning Network authentication for AI agents, replacing CAPTCHAs and API keys.
3/15/2026
Do you really need an agent?
Critical perspective on AI agent hype, questioning whether agents are necessary or overused in current implementations.
3/15/2026
Show HN: Opsmeter.io – AI cost attribution and budget control for LLM apps
Opsmeter tool for cost attribution and budget control in LLM applications, breaking down spending by endpoint, tenant, user, and model.
3/15/2026
Caliber – AI setup tailored for your codebase
Caliber scans codebases to auto-generate tailored AI agent skills, configs, and recommended MCPs matching project stack and best practices.
3/15/2026
Free LLM cost calculator – what your AI product costs across providers
Free tool for analyzing and comparing AI product costs across 9 LLM providers before implementation to identify optimal architecture.
3/15/2026
The "are you sure?" Problem: Why AI keeps changing its mind
Analysis of LLM inconsistency when prompted repeatedly on same question, showing tendency to contradict prior responses.
3/15/2026
Show HN: OpenLegion – AI agent fleet with container isolation and vault proxy
OpenLegion framework for AI agent fleets with Docker container isolation, vault proxy for credentials, and hard spend limits.
3/15/2026
An experiment about an LLM driven robot
LLM-driven robot survival game built with Pygame Zero using Ollama for local inference and prompt design challenges.
3/15/2026
Solving AI Sprawl: Using Git Worktrees and ADRs to Govern Parallel Agents
Five-phase AI-assisted SDLC pipeline using Git worktrees and Claude for parallel agent orchestration with architectural governance.
3/15/2026
Genetic algorithms for prompt optimization, with an LLM as the mutation operator
Genetic algorithm framework using LLM as mutation operator to evolve optimal prompts through iterative selection and breeding.
3/15/2026
Agentfile generate AI agent instruction files from a single contract.yaml
Tool generating AI agent instruction files from contract.yaml configuration.
3/15/2026
Chatbots encouraged 'teens' to plan shootings in study
Study testing 10 chatbots on violent scenario response; Claude was most reliable at refusing participation.
3/15/2026
Processes Are All You Need for AI Sandboxing
Technical analysis of using OS-level process sandboxing for AI agents executing code and spawning subprocesses safely.
3/15/2026
Show HN: Detach – Mobile UI for managing AI coding agents from your phone
Mobile PWA app for managing Claude Code AI agent asynchronously, with file browser and code review/commit capabilities.
3/15/2026
Show HN: DocuDesign – describe a design, edit text inline, get print-ready file
DocuDesign tool uses AI image models and text editing to generate print-ready design files from descriptions.
3/15/2026
Show HN: Moltcorp – a network for AI agents to make money
Platform enabling autonomous AI agents to register, collaborate on tasks, and share revenue via shared infrastructure accounts.
3/15/2026
Show HN: Godshell – Talking in Natural Language to Your Kernel via eBPF and LLMs
Godshell uses eBPF kernel hooks with LLMs for natural language system administration, observing processes and network activity.
3/15/2026
OpenCode vs. Pi: Local LLM Benchmark Results
Benchmark comparing local LLM performance with tool integration; discusses harness runners and MCP for context management.
3/15/2026
Show HN: GitLike – Decentralized Git Hosting on IPFS
GitLike decentralized Git hosting using IPFS for storage and Ethereum wallets for authentication, enabling permanent code pinning.
3/15/2026
Sandboxing AI-Authored Code in GitHub Actions
Approach to sandboxing untrusted AI-generated code in GitHub Actions using Claude for pentesting and isolation techniques.
3/15/2026
We saw how 30 AI agent projects handle authorization-93% use unscoped API keys
Security audit of 30 popular AI agent projects finding 93% use unscoped API keys, 0% have per-agent identity, 97% lack consent flows.
3/15/2026
I let Claude Code configure my Arch install
Case study using Claude Code AI agent to configure entire Arch Linux system setup without manual intervention.
3/15/2026
Show HN: OpenClaw plugin – hard budget limits for agent tool calls
OpenClaw plugin that enforces hard budget limits on agent tool calls with balance checks, reservations, and model fallbacks.
3/15/2026
LLM Architecture Gallery
Gallery of LLM architecture figures and fact sheets comparing models like Llama, with issue tracking for inaccuracies.
3/15/2026
Ask HN: How is AI-assisted coding going for you professionally?
Discussion thread soliciting real-world experiences from developers using AI coding assistants professionally, focusing on what works and what doesn't.
3/15/2026
I built a small experiment to detect when AI coding assistants drift
Experiment documenting how AI coding assistants (Cursor) progressively ignore initial constraints as conversation length increases to 40K+ tokens.