Live catalog
This is the browsing surface for products where agent-native behavior, agent-led creation, or agentic workflows are central. For the strategy page, read the AI agent directory. For the review standard, see how MadeWithStack evaluates agent-native products.
82
Approved listings
Products currently visible in this reviewed browse catalog.
0
Claim verified
Listings whose qualification claims have passed separate evidence review.
64
Claim submitted
Listings already in review for the agent claim trust layer.
4
Tracked stack layers
Agent tool categories with live examples attached to approved products.
Trust model
A listing can be approved for publication without automatically receiving a verified qualification claim. That keeps the public catalog useful while preserving a stricter standard for agent-directory provenance.
Approved
The product itself passed editorial review and can appear publicly.
Claim verified
The public qualification claim has supporting evidence strong enough to show as verified.
Browse reviewed products
Every filter and search state lives in the URL so the current directory view can be shared or bookmarked.
Supporting stack signals
Agent Directory listings still keep their frameworks, providers, infrastructure, and platform choices visible. That is how the catalog stays useful for technical evaluation instead of turning into a generic AI feed.
Qualification type
Verification status
Operations / Agentic workflow
Build automations with natural language not workflows
Website hosting for developers and vibe coders. Publish React, Vite, static sites and supported Next.js projects from GitHub, a CLI or an OAuth-connected MCP client, with custom domains and HTTPS.
Per-unit lien→auction compliance clock for self-storage: state-aware deadlines, proof vault, overlock status, audit trail. Free + Operator ~$29/facility/mo. Not legal advice.
Hosted AI video editor: chat edits on a real timeline, render MP4. Browser, Telegram, MCP.
Tasks, team chat, and time tracking for humans and AI agents via MCP. Free forever for unlimited people; Pro $29/mo. No per-seat pricing.
Self-hosted WordPress MCP plugin so Claude, Cursor, and ChatGPT can do real WordPress work (posts, Elementor, WooCommerce) with safety profiles — runs on your site, not a hosted chatbot.
Wrenda is a reverse proxy that sits in front of your site and detects who is requesting each page. AI crawlers get enriched markdown with schema and FAQs. Search bots get fully pre-rendered HTML. Humans get your normal page.
Search locally captured screen text and audio history for recall, meeting notes and work summaries. Connect AI agents through MCP or a local API. Optional cloud services can transmit context.
AI image, video and audio creation with a web app, API and official agent skills. Agents can prepare frames, generate media and inspect outputs. Generation uses Magic Hour credits.
Fixed-price AI memory and prompt-injection security audit for RAG, MCP, and agent teams, with direct checkout and escalation into Forge paid pilots.
Free open-source CLI for developers who want to hand off coding work too big to babysit. Ralph orchestrates Claude Code, Codex CLI, and OpenCode on your own machine so you wake up to reviewable code, checks, and artifacts.
MultiLipi is the first all-in-one platform for Multilingual SEO and Generative Engine Optimization (GEO). We help businesses rank on Google and become a cited source for AI agents like Gemini and ChatGPT in 120+ languages.
Visual AI automation and agent workflow builder
Enterprise MCP control plane for secure tool orchestration
Curated collections
Curated collection
Reviewed products where agent behavior is central to customer support, triage, or service operations.
Stack collection
Products that list Claude in their build stack or agent workflow metadata. Check each listing for supporting sources.
Checked collection
Products whose MCP endpoint passed an automated availability check: an initialize response or an OAuth sign-in requirement. Check authentication requirements and client support before connecting.