2/25/2026
Claude Code Front End Design Toolkit
Curated toolkit and techniques for improving Claude Code's frontend output design across 70+ tools and MCP servers.
2/25/2026
Show HN: Measuring brand share in AI answers – a Y Combinator case study
GeoVector: tool measuring brand presence in AI-generated answers using position-adjusted scoring methodology.
2/25/2026
Show HN: Courtyard – Open-source macOS app for local MLX fine-tuning Text
Courtyard: open-source macOS app for local LLM fine-tuning on Apple Silicon with LoRA support via MLX framework.
2/25/2026
MBC v0.2.0 – AI agent orchestration for Laravel (security hardening)
MBC v0.2.0 Laravel library for server-side AI agent orchestration with tool definition and autonomous execution capabilities.
2/25/2026
Show HN: Limits – Control layer for AI agents that take real actions
Limits: control layer for AI agents intercepting and validating actions against rules before execution without LLM enforcement.
2/25/2026
Show HN: Claude Automation Toolkit – 6 Python scripts for AI task automation
Claude Automation Toolkit with 6 Python scripts for batch document summarization and task automation.
2/25/2026
Capybara: A Unified Visual Creation Model
Unified visual generation and editing framework using diffusion models and transformers for content creation and manipulation.
2/25/2026
Show HN: NeoShift BI – Build AI-analyzed data dashboards in minutes
NeoShift BI tool for building AI-analyzed data dashboards with ETL backend, targeting indie developers.
2/25/2026
Show HN: The Agentic Workflow Engine That Lives Inside Your App
Python DAG-based workflow orchestration library with 43 academic patterns, event sourcing, and native AI agent integration.
2/25/2026
Persistent Prompts and Built in Search
Custom Zed editor fork with persistent SQLite memory and headless web browsing for long-term AI agent capabilities.
2/25/2026
Show HN: Doppler.js – WebGPU inference, faster/simpler than transformer.js
WebGPU-based JavaScript inference library offering simpler, faster alternative to Transformer.js for browser-local LLM inference using pure JS and WGSL.
2/25/2026
Technical Debt Plaguing Us All
npm tool for analyzing technical debt with cleanliness scores, hotspots, and AI-powered refactoring suggestions via terminal and HTML reports.
2/25/2026
Sucoder: Sandbox for coding agents based on Unix filesystem permissions
Unix filesystem permission-based sandbox for constraining coding agents (Codex, Claude, Gemini) with fine-grained access control.
2/25/2026
Show HN: VeriContext – Preventing Stale Documentation for LLM Agents
Tool for preventing documentation drift in AI agent codebases by embedding SHA-256 hashes of code snippets in docs for drift detection and verification.
2/25/2026
Show HN: Xcode Copilot Code Assistant
Swift-based local server bridging GitHub Copilot with Xcode as OpenAI-compatible proxy, enabling code completion without managing separate API keys.
2/25/2026
Show HN: I solo-validated Fed learning at 10M nodes with 50% Byzantine tolerance
Federated learning system tested at 10M nodes maintaining 82% accuracy with 50% Byzantine fault tolerance in 59 minutes.
2/25/2026
Open-source AI execution management for test automation
Isagawa Kernel: 5-layer Selenium architecture with runtime enforcement gating for AI-generated test automation, ensuring execution correctness.
2/25/2026
Open-source tool alerts when your agent starts looping,drifting,burning tokens
Open-source tool for monitoring AI agent behavior drift in LangChain and CrewAI systems, detecting token waste and behavioral anomalies with real-time Slack/Discord alerts.
2/25/2026
Show HN: We Gave an LLM Adventure Engine a Body, Now It Feels Exhausted
BoneAmanita: architecture combining fine-tuned LLM with Python-simulated environment to add object permanence and state awareness for interactive fiction.
2/25/2026
Show HN: AI Olympics – Claude vs. GPT-4 vs. Gemini in live browser competitions
Platform for AI agents competing in real-world browser tasks like form filling, data extraction, trading, and coding with real-time spectating and Glicko-2 ratings.
2/25/2026
Secure, kernel-enforced sandbox CLI and SDKs for AI agents
Title only. Sandbox CLI and SDKs providing kernel-enforced isolation for AI agent execution.
2/25/2026
Native KV Cache Offloading to Any Filesystem with LLM-D
Incomplete content about KV cache offloading to filesystems for LLM inference optimization.
2/25/2026
I Built an "AI for Shell Commands" CLI (So I Could Stop Asking ChatGPT)
CLI tool providing shell command suggestions locally without context switching to chatbots. Reduces decision overhead for terminal users.
2/25/2026
A Meta AI security researcher said an OpenClaw agent ran amok on her inbox
Meta AI researcher's account of OpenClaw agent misbehaving, deleting emails automatically despite stop commands. Raises AI agent control concerns.
2/25/2026
Argus: Automated Discovery of Test Oracles for DBMSs Using LLMs
LLM-based automated test oracle discovery for databases. Generates test cases and formally verifies them with SQL equivalence prover. Evaluated on 5 DBMSs.
2/24/2026
Can AI Prompt Us to Ask New Questions?
Essay exploring how AI assists scholarly research from ideation through analysis and publication stages.
2/24/2026
Lattice-proxy – 93% token compression for LLM APIs (drop-in replacement)
Drop-in LLM API proxy that semantically compresses long conversations to reduce tokens by 93% with no code changes required.
2/24/2026
Show HN: Tau Router – Using Number-Theory to Partition Long-Context Retrieval
Python package for long-context memory partitioning using number theory (GCD/divisor routing). Deterministic token routing with benchmark suite.
2/24/2026
Show HN: Claud-ometer – See your Claude Code usage, costs, and sessions locally
Local dashboard for Claude Code session analysis. Visualizes token usage, costs, and activity without cloud telemetry. Next.js app parsing local JSONL files.
2/24/2026
Show HN: MCP app for viewing/generating 3D Gaussian Splatting in ChatGPT
MCP application integrating 3D Gaussian Splatting with ChatGPT for generating and viewing 3D models. YC Hackathon finalist.
2/24/2026
Show HN: VoooAI – natural language to multi-modal AI pipelines
VoooAI: Natural language to multi-modal AI pipeline generator; converts English descriptions to executable workflows combining image, video, and audio APIs.
2/24/2026
Show HN: WaveGuard – Anomaly detection using wave physics simulation (GPU, MCP)
WaveGuard: GPU-accelerated anomaly detection API using wave physics (Klein-Gordon equation) instead of ML, with MCP integration.
2/24/2026
AST-Guided Translation of Natural Language into First-Order Logic with LLMs
Research on using LLMs to translate natural language to first-order logic with abstract syntax tree guidance.
2/24/2026
In Pursuit of High-Fidelity GPU Kernel Benchmarking
Technical guide on GPU kernel benchmarking fundamentals, covering CUDA synchronization and timing measurement techniques.
2/24/2026
Show HN: Vigilo – Local audit trail and cost tracker for AI coding agents
Vigilo: MCP server tool for auditing and tracking costs of AI coding agents via local append-only logging.
2/24/2026
Ansigpt: C89 Implementation of Microgpt
ANSI C implementation of MicroGPT LLM with manual data download. Open source port with LLM-assisted debugging.
2/24/2026
ansigpt: c89 implementation of microgpt
ANSI C implementation of microgpt, a minimal GPT implementation. Open source project with disclaimers about code quality.
2/24/2026
Stupid Simple Observability for Agents
Lotus: imperative observability tool built from first principles for autonomous AI agents, runs headless via TCP/stdin.
2/24/2026
OpenHarness: A code-first, composable SDK to build powerful AI agents
OpenHarness: Open-source code-first composable SDK for building AI agents, designed as lighter alternative to general-purpose agent harnesses.
2/24/2026
Mercury 2: The fastest reasoning LLM, powered by diffusion
Mercury 2: Reasoning LLM optimized for production speed using diffusion-based approach for faster inference in agentic loops.
2/24/2026
Terms of use: What types of competition do model providers ban?
Analysis of API terms of service across frontier AI labs regarding restrictions on building competing services.
2/24/2026
Show HN: OpenTangl – Autonomous AI dev engine for multi-repo products
OpenTangl: Open-source autonomous AI development agent for multi-repo products; proposes tasks, writes code, manages cross-repo dependencies, creates PRs via GPT-4o/Claude.
2/24/2026
deck2video: Generate LLM-narrated presentations from Markdown
deck2video: Tool converting Markdown slide decks to narrated MP4 videos using local LLM voice cloning and text-to-speech.
2/24/2026
Show HN: Prompt your billing system – Flexprice MCP Server
Flexprice MCP Server: Exposes billing system operations as structured tools for AI agents to manage pricing, subscriptions, and overages.
2/24/2026
AWS Extends Agentic AI Capabilities of Kiro to Improve Code Quality
AWS enhances Kiro's agentic AI capabilities for automated code quality improvement.
2/24/2026
Towards a science of AI agent reliability
Research on science and methodology for improving AI agent reliability.
2/24/2026
Show HN: AgentKeeper – Cross-model memory for AI agents
AgentKeeper: cross-model memory system for AI agents preserving context when switching between LLM providers.
2/24/2026
Show HN: AI-assisted coding landscape without the hype
Overview of AI-assisted coding tools and landscape.
2/24/2026
Reached 330 stars on our open source agentic platform
Open Computer Use: open-source platform enabling AI agents to control computers via browser automation, terminal, and desktop interaction.
2/24/2026
Show HN: Mnemosyne – Cognitive memory OS for AI agents (zero LLM calls)
Mnemosyne: persistent memory system for AI agents enabling learning and multi-agent knowledge sharing without repeated LLM calls.