The datasheet for every AI agent

AI-agent reference · rev 2026

Every AI agent, on one datasheet.

What each one is made of, what it costs, and whether it holds under load. Every part researched, spec'd, sourced and dated, the reference you check before you wire an agent into a system.

The market, measured · last verified 27 Aug 2026

The Agents Index tracks 95 agents across 9 categories, each researched from primary sources and re-verified on a published cadence.

The index

Sort the index
How we rankBrowse all agents

This week's picks — agents. Column headers sort.
Best for
JunieCoding agentsTeams already standardized on JetBrains IDEs (IntelliJ, PyCharm, WebStorm, GoLand, PhpStorm, Rider, RubyMine, RustRover) who want an agent driving the same debugger and project index they already use.Free / from $10/mo
AgentsKitAgent frameworksDevelopers building custom agents in Node who prefer copying owned, typed source over subscribing to a hosted agent platform.On quote
Muse CodeCoding agentsMeta's terminal coding agent — pay standard rate and it never trains on your code, or opt into Contributor pricing for a steep discount if you let it.Usage-based
WebhoundResearch & data agentsTechnical buyers and developers who want a research agent callable from inside Claude, Codex or Cursor via MCP, or who need a sourced Dataset rather than a narrative Report.$1 one-time
AgnoAgent frameworksAn Apache 2.0 Python agent SDK and AgentOS runtime, with a Control Plane you run in your own cloud.Free / from $150/mo
CartesiaVoice agentsThe Stanford SSM-research spinout behind Sonic (TTS) and Ink (STT) — Line, its open-source voice-agent SDK, connects to 100+ LLMs and starts at a genuine $0/month, but the agent layer itself isn't self-hostable.Free / from $5/mo
Deepgram Voice Agent APIVoice agentsEngineering teams that need managed real-time STT plus TTS plus LLM orchestration over a single WebSocket, with the broadest BYO-LLM router in the category and a real self-hosted or VPC deployment option.$4.50
OpenAI Realtime APIVoice agentsOpenAI's own speech-to-speech model API for voice agents: WebRTC, WebSocket or native SIP telephony, billed purely on audio tokens with no platform fee.Usage-based
Lead ScorerSales & marketing agentsA solo founder or small B2B team who wants an autonomous outbound SDR they can drive from Claude or ChatGPT, with French registry sourcing and human approval on every send.Free / from €99/mo
VoiceAgentVoice agentsNigerian SMB and mid-market operators whose callers speak Yoruba, Igbo, Pidgin, Hausa or Nigerian English and who want a no-code phone agent live in minutes.$12.50/mo

Newest — agents. Column headers sort.
Best for
Lead ScorerSales & marketing agentsA solo founder or small B2B team who wants an autonomous outbound SDR they can drive from Claude or ChatGPT, with French registry sourcing and human approval on every send.Free / from €99/mo
VoiceAgentVoice agentsNigerian SMB and mid-market operators whose callers speak Yoruba, Igbo, Pidgin, Hausa or Nigerian English and who want a no-code phone agent live in minutes.$12.50/mo
Agentix LabsAgencies & build partnersOperations and IT leaders in Canada or North America who want a partner to design, build, and operate governed AI agents in production.On quote
VellumAgent platformsIndividuals who want a memory-persistent personal assistant that can act across email, calendar, chat, and phone with visible permission checkpointsFree / from $30/mo
SimAgent platformsEngineering teams that want an open-source, self-hostable workspace to build and deploy multi-app AI agent workflowsFree / from $25/seat/mo
MindStudioAgent platformsNon-technical teams and business units automating multi-step workflows with pre-built integrations and templatesFree / from $20/mo
Kilo CodeCoding agentsDevelopers and teams who want an open source, model-agnostic AI coding agent across IDE, CLI, and cloudFree / from $15/seat/mo
GooseCoding agentsDevelopers and engineering teams who want a free, self-hosted coding agent they can point at any LLM provider and extend with MCP toolsFree
TinyFishResearch & data agentsEngineering and growth teams that need programmatic web data or browser automation without building and maintaining their own scraping infrastructure.Freemium
BrowserbaseAgent tools & infrastructureEngineering teams building production browser-automation or AI-agent products who want managed cloud browsers, session recording, and a broad API surface instead of self-hosting Playwright infrastructure.Free / from $20/mo

Recently verified — agents. Column headers sort.
Best for
Bland AIVoice agentsEnterprises in healthcare, insurance, financial services and logistics with engineering capacity who need an owned-stack voice platform that meets compliance review on day one.today
VecbaseAgent platformsSmall teams or solo operators who want pre-built AI agent roles, shared memory and files, and bundled capabilities like web search, document parsing, and image generation under one credit-based subscription.today
Lead ScorerSales & marketing agentsA solo founder or small B2B team who wants an autonomous outbound SDR they can drive from Claude or ChatGPT, with French registry sourcing and human approval on every send.today
Synthflow AIVoice agentsA no-code voice AI platform for enterprise phone agents — a visual Flow Designer and a structured Build-Evaluate-Launch-Learn deployment cycle, billed by the minute in separate components.2 days ago
CursorCoding agentsThe AI-first code editor (a VS Code fork) with a coding agent that reads, plans and edits across your whole codebase. Free Hobby tier, Pro from $20/month.2 days ago
Bolt.newCoding agentsBolt.new is StackBlitz's browser-native prompt-to-full-stack-app builder. WebContainers runs a real Node environment in the browser tab itself, no remote VM, no local setup.2 days ago
VoiceAgentVoice agentsNigerian SMB and mid-market operators whose callers speak Yoruba, Igbo, Pidgin, Hausa or Nigerian English and who want a no-code phone agent live in minutes.3 days ago
Replit AgentCoding agentsSolo builders and small teams who want one platform to go from prompt to hosted app, including database, auth and deployment.3 days ago
Microsoft Agent FrameworkAgent frameworks.NET or Python teams standardizing on Azure or Microsoft Foundry who want one supported agent SDK with first-party migration paths off AutoGen and Semantic Kernel.3 days ago
LindyAgent platformsTeams that want an AI teammate embedded in Slack, with the no-code Lindy Build product and iMessage 24/7 access bundled in on every plan. Especially useful for sales, support, and ops leads who need inbox and calendar handling end to end, and for non-technical operators who want to ship internal tools from a plain-language prompt.3 days ago

Browse the index

Search all agents

Have an agent people should find?

Submitting is free. Claim your listing to keep its facts fresh. Every listing passes the same quality gate; paying never skips it.

Submit an agent, free

Want to reach the people reading it?

Sponsored placements reach developers and AI-tool buyers, plus every major answer engine. Always labelled, never mixed into the editorial listings.

See placements & pricing

Frequently asked questions

How do agents get listed on The Agents Index?

Each listing is a researched record that must pass an automated quality gate (a minimum of structured fields, sourced facts and an on-topic fit-check) before it can be published. Thin pages are structurally impossible.

Is The Agents Index free to use?

Yes. Browsing and searching the directory is free, and submitting an agent is free too. Every submission is researched and quality-gated the same way.

Can an agent pay for a better rank?

No. Rankings and verdicts are editorial and the quality gate is never for sale. What makers can buy is clearly-labelled placement (Featured spots and sponsorships), which never changes what our research says about them.

How fresh is the data?

Every listing is re-researched on a schedule, and each fact carries the source it was verified from, with the verification date printed on the listing. Changes land in the listing changelog, not silently.

Do you rate or score agents?

No. A rating is only worth reading when its input is independent of whoever publishes it — thousands of verified users, or a product measured on a bench. We do research instead, and any number derived from research would be measuring how much we found, not how good the agent is. So every listing lists the sources it is built from — each labelled with the claims it supports — says when we last re-verified it and what changed, names who it is best for and who it is not for, and lists real limitations.

I built an agent. How do I get it covered?

Submit it free. The research agent builds a sourced listing, the quality gate checks it, and if it fits the taxonomy it publishes. You claim it from your inbox and manage it from your dashboard.