All AI Tools
Hand-reviewed. Every tool verified by a real person before it's listed.
Cursor
Score
Cursor is a standalone AI-powered code editor available for Windows, macOS, and Linux, developed by Anysphere. It embeds deep codebase awareness, natural-language editing, and multi-file refactoring into the editor. Users can autocomplete code with Tab, ask questions or apply changes via chat, and run terminal commands through agent workflows. Cursor supports a range of frontier models including GPT-4, Claude, and Gemini, and offers privacy controls like SOC-2 certification and Privacy Mode. It’s aimed at developers who want AI tightly integrated into their coding workflow.
Roblox GUI Maker
Score
Roblox GUI Maker is a web-based tool that converts simple text prompts into visual GUI mockups for Roblox Studio. It offers a visual editor with drag-and-drop support for elements like frames, buttons, labels, and scrolling areas, and lets you preview layouts on desktop and mobile. The tool exports clean Luau code - complete with Instance.new calls, layout constraints, and optional button wiring - that you can paste directly into Roblox Studio. It supports starting from templates such as shops, HUDs, menus, and reward popups, and runs entirely in the browser without requiring an account.
Tura
Score
Tura is a local, open-source coding agent for developers who are tired of vague skill claims, token-saving extensions with no evidence, and agents that change a repository before understanding it. Tura benchmarks its work and reports 16.7% better performance with 77.5% fewer rounds. Source: https://github.com/Tura-AI/tura. Reproducible benchmark: https://turaai.net/benchmark. I am affiliated with the Tura project, and AI assistance was used to prepare and verify this submission.
aicoolies
Score
Hand-tested knowledge graph of AI developer tools - curated comparisons and research for developers choosing AI coding tools.
Pangolinfo Amazon Data MCP
Score
Amazon Data MCP connects MCP-compatible AI assistants to structured Amazon product, search, review, seller, category, ranking and niche data. Developers can use the hosted endpoint or the open-source stdio bridge to build e-commerce research workflows. The 19 tools are read-only. The bridge is free; the backend requires a Pangolinfo API key, with trial access and paid usage limits. Source and setup instructions: https://github.com/Pangolin-spg/amazon-data-mcp
API.market
Score
API.market is a hosted API marketplace that lets developers discover, subscribe to, and call APIs across categories including image generation, video, maps, aviation data, face-swap tools, text analysis, and more. The platform lists 590+ APIs and AI models from independent sellers, with each seller setting their own pricing, free trial units, and usage limits. Beyond the marketplace, API.market operates a hosted MCP (Model Context Protocol) gateway that gives AI agents a single connection point to over 3,650 tools drawn from the catalog. Agents connect via Streamable HTTP using OAuth or an API key, then search the catalog, retrieve schemas, make API calls, and manage subscriptions programmatically. An API.market account is required to use either the marketplace or the gateway.
Qveris
Score
QVeris is a capability routing network that gives AI agents structured access to over 10,000 real-world APIs, live data sources, and external services through a unified protocol. Developers integrate via a CLI, Python SDK, JavaScript SDK, MCP server, or REST API. The platform follows a Discover-Inspect-Call model: agents search the capability catalog in natural language, review parameter schemas and estimated credit costs, then execute calls that return structured JSON output with a full audit trail. The platform has a particular depth in financial data, covering equities, fixed income, FX, crypto, macroeconomic indicators, and alternative signals across 138 documented finance capabilities. Pricing is pay-as-you-go with no monthly subscription: new accounts receive 1,000 free credits, and purchased credits never expire. QVeris targets individual developers and teams building production AI agents that need auditable, cost-predictable access to external tools and data.
ScrapeAtlas
Score
ScrapeAtlas is a unified social media data API for public profiles, posts, videos, and conversations across ~40 platforms. REST with x-api-key (https://api.scrapeatlas.com); hosted MCP Streamable HTTP with Bearer (https://api.scrapeatlas.com/mcp) so AI agents can pull social data directly. Docs: https://scrapeatlas.com/mcp/. Freemium: 100 free credits, no card. Company: OD. Media & Technology LLC.
RAG Readiness Checklist (Free)
Score
Sandesh Kale (Singapore), GenAI/RAG practitioner. Free Gumroad download, not a SaaS app, chatbot, scoring engine, OCR service, vector DB, certification, automated tester, or a guarantee of answer quality. Price: Free ($0). Pay-what-you-want is on, so Gumroad shows $0+. Downloads usually collect an email. Tool-neutral checklist for planning or reviewing a RAG app: data, permissions, freshness, ownership; parsing, chunking, metadata, embeddings, retrieval; citations, refusal, prompt injection, guardrails; evaluation and regression checks; model, latency, token cost, failure handling; logs, traces, feedback, incidents, ops; privacy, secrets, retention, access, deployment; a ship-vs-needs-evidence decision. Update policy: I revise this checklist when major RAG practices change and re-upload the file on Gumroad. Optional next step, not required: n8n PDF RAG + Eval Starter (sandeshkale.gumroad.com/l/fhfzp). Related: Production RAG + n8n Starter Pack (sandeshkale.gumroad.com/l/fombgg).
Poket Dev
Score
Poket Dev is a subscription-based software development service where clients submit unlimited requests to a private Trello board and receive code delivered to a private GitHub repository, one active request at a time. The founder personally builds every project, targeting a 72-hour turnaround per request with unlimited revisions until the client approves. The service covers Python-first development including AI and ML integrations, Django and FastAPI backends, automation, data engineering, web scraping, and cloud infrastructure on AWS, GCP, and Azure, with JavaScript and TypeScript available for full-stack work. Plans are priced at $5,995 per month on a monthly basis, $5,495 per month on a quarterly basis, and $4,995 per month on an annual basis, with no upfront fees and cancel-anytime terms. The service targets startup founders, small teams, and companies that need ongoing custom software built without the cost or commitment of a full-time engineering hire. All code produced remains the client's property.
Modurian
Score
Modurian is a web-based multi-agent AI workspace designed for developers and founders who want to build and deploy web applications from a plain-language description. It coordinates specialized agents across three modes: Plan mode drafts a structured plan before touching code, Build mode writes files and runs npm in a sandboxed environment with automatic error correction, and Ask mode answers questions without modifying any files. Completed projects can be deployed to shareable Spaces URLs, exported as a ZIP, or pushed to GitHub, all without configuring API keys. The platform uses automatic model routing to select the appropriate AI model for each task. The Pro tier enables Snap-to-Build (screenshot-to-code), vision uploads, a prompt optimizer, message checkpoints with session restore, and the full skills and plugins library. Modurian targets solo builders and early-stage founders who want a fast path from concept to a publicly accessible URL without managing infrastructure.
CloudPloy
Score
CloudPloy is a cloud deployment platform designed around the Model Context Protocol, letting developers deploy and manage applications directly from AI coding tools like Claude Code and Cursor. It acts as a control plane that connects to servers on Hetzner, DigitalOcean, or AWS, either ones the user already owns or ones CloudPloy provisions on demand at the provider's exact cost. The platform targets individual developers and small engineering teams who want AI agents to handle deployment tasks without custom scripting. A permanent free tier supports one server and one app within a 2 vCPU and 2 GB RAM ceiling. Paid plans start at $9.99 per month and raise server count, resource limits, and team seat allowances, while compute costs remain separate and unmarketed.
The Agentic Coding Kit
Score
The Agentic Coding Kit is a drop-in configuration pack that makes AI coding assistants follow senior-engineer discipline from the first prompt. It ships CLAUDE.md presets for five stacks (Python/FastAPI, TypeScript/Node, React/Next.js, Go, Rust), six Cursor rules for testing, git, security and style, five Claude Code subagents (reviewer, test writer, security auditor, refactorer, docs writer), four slash commands (/review, /test, /secure, /ship), git hook recipes, CI templates, pre-commit config and ADR templates. Plain text files, no lock-in. One-time purchase, free v1.x updates.
MCPGRAM
Score
MCPGRAM is a cloud-hosted MCP gateway that centralizes tool access for AI coding agents. Developers connect OAuth-authenticated apps such as GitHub, Slack, Notion, and Linear alongside their own external MCP servers in a single workspace, and every supported agent client, including Claude Code, Cursor, Codex, Windsurf, and Cline, reads from that shared tool catalog instead of requiring separate per-client configuration. The platform is accessible via a web dashboard, a CLI published on npm as @mcpgram/cli, and a JS/TS SDK. Launched in July 2026 by an independent founder, MCPGRAM targets individual developers and small engineering teams who manage multiple AI agent clients and want centralized OAuth governance, per-workspace toggle controls, and usage visibility without maintaining separate credential stores per tool. The free Hobby plan requires no credit card and includes 25K tool runs per month; paid plans extend run limits, seat counts, and workspace quotas.
CodeGuard AI
Score
CodeGuard AI is a self-hosted AI code review application for GitHub pull requests. It reviews pull requests using repository context and configurable team rules instead of relying only on the changed diff. It can identify bugs, security issues, performance problems, and coding concerns. The application also provides behavior impact analysis, regression scenarios, AI-generated fix suggestions, patch validation, reviewer feedback learning, GitHub integration, and review history. CodeGuard AI supports OpenAI, Google Gemini, and Ollama. It is provided as source code so developers can self-host and customize the review workflow.