Evaluate B2B software vendors via AI agent conversations with evidence-based scoring across 7 dimensions.
Drive OpenAI Codex from Claude Code — run tasks, get code reviews, and do parallel research.
Audit designs against 19 professional design rules — accessibility, ethics, dark patterns, and more.
Guide AI agents through architecture review, cost modeling, PM planning, and anti-pattern detection for regulated software delivery.
Transform codebases into literate programs with prose, diagrams, and code chunks for Claude Code.
Control VMware vSphere VMs, clusters, and deployments via AI agents.
Analyze images, websites, and Figma files into structured design specs with DTCG tokens and rebuild prompts.
Generate ranked research directions daily from your papers and fresh literature.
Benchmark and evaluate agent skills with deterministic evals across LLMs.
Voice-enable LLM agents with local speech-to-text and text-to-speech via Kesha Voice Kit.
Guide family history research with GPS standards, citation templates, and evidence analysis.
Reflective practice skill for AI agents — check intentions, capture learnings, notice breakdowns before, during, and after action.
Updates terminal window titles based on Claude Code tasks so you never lose track of which terminal is which.
Generate pairwise combinatorial test cases from requirements using PICT methodology.
Generate FLUX images via prompting best practices and BFL API integration patterns.
Mine repeated coding-agent workflows, personalize skills, and generalize them for public release.
Make your AI assistant fluent in Longbridge — stock prices, portfolio, news, valuations in plain English or Chinese.
Complete Kaggle integration for AI coding agents — competitions, datasets, notebooks, badges, and hackathon writeups.
Self-improvement system for Hermes Agent. Monitors performance, finds weak skills, fixes them with self-evolution, reports results.
Drive Git and GitHub workflows for Claude Code — structured commits, PRs, reviews, and merges with safety guardrails.
Strip AI-isms from assistant output to sound human across 6 coding platforms.
Generate design specs, UI reviews, and UX guidelines through structured interviews for AI coding agents.
Turn accumulated sources into a self-maintaining markdown knowledge base via LLM.
Browse, install, and verify 2,600+ reusable AI agent skills for OpenClaw, Claude Code, Cursor, and MCP.
Evolve LLM agent prompts, routing, and tools via multi-agent evolution with LangSmith evaluation.
210 evaluated academic Claude/agent skills across 16 research domains — bioinformatics, data science, Turkish academia & more.
Query data and build semantic models through natural conversation with coding agents.
Stop AI coding assistants from over-engineering code with minimal-change rules.
Turn AI coding sessions into animated, interactive web replays.
Swift server development skills with automated linting for Claude Code.
