Find your next agent. Compare the details.

1,752 agents, frameworks, and tools. Search for a project, browse by task, or compare up to four entries.

What the coverage labels mean

Product docs checked means we have a dated review of official sources; open an entry’s review scope to see what was checked. It does not mean hands-on testing or that every attribute is known. Catalog entries have spec sheets; discovered entries await individual review. Workflow groupings are Agentlist’s editorial organization, assessed separately from product documentation and marked when provisional. Hosting and interface filters include only documented values, using the same evidence as the comparison table.

Reviewed categories have dedicated fields; entries awaiting classification share only common attributes.

All categories

1,752 entries, page 1 of 59

Select entries to build a comparison.

  • Cursor

    Catalog entry · check review scope

    Cursor is a VS Code–based AI-native editor with repo-aware chat, multi-file edits, Composer/agent modes, and cloud background agents.

    productCatalog entry
    Sources & review scope

    Agentlist workflow assignment: Provisional grouping from the recorded description. The description identifies a coding assistant or live editor/terminal interaction as its main interface. Product documentation has not been rechecked for this grouping. Provisional; individual workflow review is pending.

    Historical catalog snapshot; no individual verification date recorded. (2026-09-25)

    Visit source
  • Claude Code

    Product docs checked2026-11-05

    Claude Code is Anthropic's terminal-native coding agent: it plans, edits, runs commands, uses Git and MCP, and sustains long agentic loops.

    productCatalog entry
    Sources & review scope · 2026-11-05

    Agentlist workflow assignment: Provisional grouping from the recorded description. The description identifies a coding assistant or live editor/terminal interaction as its main interface. Product documentation has not been rechecked for this grouping. Provisional; individual workflow review is pending.

    Historical catalog snapshot; no individual verification date recorded. (2026-09-25)

    Official source checked for product identity and scope; no performance or current price verification. (2026-11-05)

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • Codex CLI

    Product docs checked2026-11-05

    Codex is OpenAI's coding agent: an open-source terminal CLI, a desktop app, IDE extensions, and a cloud lane for async tasks, all included in ChatGPT plans.

    productCatalog entry
    Sources & review scope · 2026-11-05

    Agentlist workflow assignment: Provisional grouping from the recorded description. The description names a coding agent, consistent with the recorded interactive-coding role. Product documentation has not been rechecked for this grouping. Provisional; individual workflow review is pending.

    Historical catalog snapshot; no individual verification date recorded. (2026-09-25)

    Official source checked for product identity and scope; no performance or current price verification. (2026-11-05)

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • Sierra

    Catalog entry · check review scope

    Sierra is the customer-service AI employee for large consumer brands: it resolves support conversations across chat and voice, takes actions in business systems, and bills on outcomes.

    productCatalog entry
    Sources & review scope

    Historical catalog snapshot; no individual verification date recorded. (2026-09-25)

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • Fin

    Product docs checked2026-11-10

    Fin is Salesforce’s customer support agent, formerly from Intercom. It answers from your support content, resolves conversations, and hands off to a human when needed.

    productCatalog entry
    Sources & review scope · 2026-11-10

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • OpenCode

    Catalog entry · check review scope

    OpenCode is an MIT-licensed open-source terminal coding agent that is fully BYOK, with optional hosted model access via Zen and Go.

    productCatalog entry
    Sources & review scope

    Agentlist workflow assignment: Provisional grouping from the recorded description. The description identifies a coding assistant or live editor/terminal interaction as its main interface. Product documentation has not been rechecked for this grouping. Provisional; individual workflow review is pending.

    Historical catalog snapshot; no individual verification date recorded. (2026-09-25)

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • Grok Bot

    Product docs checked2026-11-10

    Grok Bot is an AI teammate with a persistent cloud computer. It works across connected apps and websites, continues tasks in the background, and can ask for approval before taking an action.

    productCatalog entry
    Sources & review scope · 2026-11-10

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • v0

    Product docs checked2026-11-05

    App and interface creation with deployment and PR workflows.

    productSource-checked
    Sources & review scope · 2026-11-05

    Agentlist workflow assignment: Provisional grouping from the recorded description. The main description names a development task or application build that the reader hands over for a result. Product documentation has not been rechecked for this grouping. Provisional; individual workflow review is pending.

    Official source checked for product identity and scope; no performance or current price verification. (2026-11-05)

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • Muse

    Product docs checked2026-11-10

    Muse is Meta’s personal AI agent for tasks like sorting email, managing calendars, and booking reservations. It can continue working after you close the app and asks for approval for actions such as sending an email or making a purchase.

    productCatalog entry
    Sources & review scope · 2026-11-10
    Visit source
  • Harvey

    Catalog entry · check review scope

    Harvey is the AI associate for legal work: drafting, review, diligence, and research grounded in legal corpora, deployed inside law firms and enterprise legal departments.

    productCatalog entry
    Sources & review scope

    Historical catalog snapshot; no individual verification date recorded. (2026-09-25)

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • Devin

    Catalog entry · check review scope

    Devin is Cognition's autonomous software engineer: cloud sandboxes for ticket-in / PR-out work, plus Devin Desktop and CLI on the same plan as Windsurf.

    productCatalog entry
    Sources & review scope

    Agentlist workflow assignment: Provisional grouping from the recorded description. The main description names a development task or application build that the reader hands over for a result. Product documentation has not been rechecked for this grouping. Provisional; individual workflow review is pending.

    Historical catalog snapshot; no individual verification date recorded. (2026-09-25)

    Visit source
  • Salesforce Agentforce

    Product docs checked2026-11-10

    Salesforce Agentforce is a platform for building and deploying agents that work with Salesforce data and business workflows. Teams configure what each agent can do through actions, instructions, and permissions.

    productCatalog entry
    Sources & review scope · 2026-11-10

    Official source checked for product identity and scope; no performance or current price verification. (2026-11-05)

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • Glean Agents

    Product docs checked2026-11-10

    Glean Agents lets teams build agents that work with company knowledge and connected business apps. Teams can create workflows in natural language and use enterprise access controls to govern what an agent can read and do.

    productCatalog entry
    Sources & review scope · 2026-11-10

    Official source checked for product identity and scope; no performance or current price verification. (2026-11-05)

    Visit source
  • n8n

    Product docs checked2026-11-10

    n8n is a visual workflow automation platform. Its AI Agent node can choose among connected tools inside a workflow, and teams can run n8n in the cloud or self-host it.

    productCatalog entry
    Sources & review scope · 2026-11-10

    Official source checked for product identity and scope; no performance or current price verification. (2026-11-05)

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • Google Antigravity

    Catalog entry · check review scope

    Google Antigravity is Google's agent-first IDE and CLI, the successor surface to Gemini Code Assist for individuals and Gemini CLI.

    productCatalog entry
    Sources & review scope

    Agentlist workflow assignment: Provisional grouping from the recorded description. The description identifies a coding assistant or live editor/terminal interaction as its main interface. Product documentation has not been rechecked for this grouping. Provisional; individual workflow review is pending.

    Historical catalog snapshot; no individual verification date recorded. (2026-09-25)

    Visit source
  • Decagon

    Catalog entry · check review scope

    Decagon builds autonomous customer-support agents that resolve chats, emails, and calls for mid-market and enterprise brands, with engineering-grade tooling for the ops team behind them.

    productCatalog entry
    Sources & review scope

    Historical catalog snapshot; no individual verification date recorded. (2026-09-25)

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • ElevenLabs Agents

    Catalog entry · check review scope

    ElevenLabs Agents is the voice agent platform built on the industry's best-known TTS: conversational agents with top-tier voices, phone and web deployment, and model-agnostic reasoning.

    productCatalog entry
    Sources & review scope

    Historical catalog snapshot; no individual verification date recorded. (2026-09-25)

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • Muse Code

    Catalog entry · check review scope

    Muse Code is Meta's terminal coding agent powered by Muse Spark 1.2 — persistent background subagents, a replay-exact event log, workflow orchestration, and a TypeScript SDK over the Muse Session Protocol.

    frameworkCatalog entry
    Sources & review scope

    Agentlist workflow assignment: Provisional grouping from the recorded description. The description identifies a coding assistant or live editor/terminal interaction as its main interface. Product documentation has not been rechecked for this grouping. Provisional; individual workflow review is pending.

    Historical catalog snapshot; no individual verification date recorded. (2026-09-25)

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • OpenRouter

    Catalog entry · check review scope

    OpenRouter is a single API to many LLM providers with routing, fallbacks, and usage billing — widely wired into Aider, Continue, Goose, and OpenHands.

    infrastructureCatalog entry
    Sources & review scope

    Historical catalog snapshot; no individual verification date recorded. (2026-09-25)

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • ChatGPT Deep Research

    Catalog entry · check review scope

    ChatGPT Deep Research is OpenAI's research-agent mode: a prompt becomes a 5–30 minute browsing run that returns a cited report — now also reachable via API as a deep-research model.

    infrastructureCatalog entry
    Sources & review scope

    Historical catalog snapshot; no individual verification date recorded. (2026-09-25)

    Visit source
  • Browserbase

    Product docs checked2026-11-05

    Browserbase is browsers-as-a-service for web agents — hosted headless browsers with stealth, proxies, sessions, and replay, the runtime underneath its own Stagehand framework and third-party agents.

    infrastructureCatalog entry
    Sources & review scope · 2026-11-05

    Historical catalog snapshot; no individual verification date recorded. (2026-09-25)

    Official source checked for product identity and scope; no performance or current price verification. (2026-11-05)

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • Exa

    Catalog entry · check review scope

    Exa is a search engine built for agents — an index queried by meaning (neural + keyword) with APIs that return clean page contents instead of a page of links.

    infrastructureCatalog entry
    Sources & review scope

    Historical catalog snapshot; no individual verification date recorded. (2026-09-25)

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • E2B

    Product docs checked2026-11-05

    E2B provides secure sandboxes for agent-generated code — instant isolated VMs where an agent can run untrusted code, install packages, and keep files, widely used as the compute layer behind AI apps.

    infrastructureCatalog entry
    Sources & review scope · 2026-11-05

    Agentlist workflow assignment: Provisional grouping from the recorded description. The recorded developer infrastructure helps build, run or evaluate agent applications; it is adjacent to coding workspaces. Product documentation has not been rechecked for this grouping. Provisional; individual workflow review is pending.

    Historical catalog snapshot; no individual verification date recorded. (2026-09-25)

    Official source checked for product identity and scope; no performance or current price verification. (2026-11-05)

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • AlphaSense

    Catalog entry · check review scope

    AlphaSense is the institutional market-intelligence platform whose Deep Research agents search premium content — broker research, filings, transcripts, expert calls — and return decision-grade cited reports.

    productCatalog entry
    Sources & review scope

    Historical catalog snapshot; no individual verification date recorded. (2026-09-25)

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • Cline

    Catalog entry · check review scope

    Cline is an open-source VS Code agent that plans, edits, and runs tools with BYOK models and approval controls.

    productCatalog entry
    Sources & review scope

    Agentlist workflow assignment: Provisional grouping from the recorded description. The description identifies a coding assistant or live editor/terminal interaction as its main interface. Product documentation has not been rechecked for this grouping. Provisional; individual workflow review is pending.

    Historical catalog snapshot; no individual verification date recorded. (2026-09-25)

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • CodeRabbit

    Product docs checked2026-11-05

    Code review and change-management platform.

    productSource-checked
    Sources & review scope · 2026-11-05

    Agentlist workflow assignment: Provisional grouping from the recorded description. The description centers on automated code-review findings; reviewing changes within another tool is not this job. Product documentation has not been rechecked for this grouping. Provisional; individual workflow review is pending.

    Official source checked for product identity and scope; no performance or current price verification. (2026-11-05)

    Identity also present in the historical discovery inventory; not a new product review. (2026-09-25)

    Visit source
  • ChatGPT Work

    Product docs checked2026-11-05

    Outcome-oriented knowledge-work experience.

    productSource-checked
    Sources & review scope · 2026-11-05

    Official source checked for product identity and scope; no performance or current price verification. (2026-11-05)

    Visit source
  • Claude knowledge work / Cowork

    Product docs checked2026-11-05

    General knowledge-work agent across files and connected tools.

    productSource-checked
    Sources & review scope · 2026-11-05

    Official source checked for product identity and scope; no performance or current price verification. (2026-11-05)

    Visit source
  • Claude Agent SDK

    Product docs checked2026-11-05

    Programmable Claude Code agent loop in Python and TypeScript.

    frameworkSource-checked
    Sources & review scope · 2026-11-05

    Agentlist workflow assignment: Provisional grouping from the recorded description. The recorded product is a programmable coding-agent loop for developers to embed. Product documentation has not been rechecked for this grouping. Provisional; individual workflow review is pending.

    Official source checked for product identity and scope; no performance or current price verification. (2026-11-05)

    Visit source
  • Zapier Agents

    Product docs checked2026-11-05

    Agents that work with connected applications and Zapier automation.

    productSource-checked
    Sources & review scope · 2026-11-05

    Official source checked for product identity and scope; no performance or current price verification. (2026-11-05)

    Visit source

Curated comparison guides

26 curated pairings — spec fields side by side, plus pick-which guidance. Documentation-based records, reviewed individually.

CursorvsWindsurf / Devin Desktop

Anysphere · ai-ide / Cognition · ai-ide

The two flagship agentic development environments. Cursor is a VS Code fork with the largest third-party model pool; Windsurf (now under Cognition, alongside Devin) is the ex-Codeium editor built around Cascade agent flows.

Claude CodevsCodex CLI

Anthropic · cli-agent / OpenAI · cli-agent

The two terminal-native agents bundled with the frontier subscriptions — Claude Pro/Max vs ChatGPT Plus/Pro. Both plan, run commands, and commit against your local checkout.

Claude CodevsCursor

Anthropic · cli-agent / Anysphere · ai-ide

Terminal agent vs agentic development environment — the two most-discussed paid setups. Claude Code runs supervised tasks from the shell; Cursor keeps the agent inside a VS Code-style editor.

GitHub CopilotvsCursor

GitHub · ide-extension / Anysphere · ai-ide

Extension vs fork. GitHub Copilot rides inside the IDEs you already use; Cursor asks you to switch editors for a deeper agent integration.

AidervsClaude Code

Community / OSS · cli-agent / Anthropic · cli-agent

The original git-native pair-programmer vs the bundled frontier CLI. Aider is open source and BYOK; Claude Code is closed, subscription-gated, and ships first-party models.

ClinevsRoo Code

Community / OSS · ide-extension / Roo Code Inc. · ide-extension

The two best-known OSS VS Code agents. Cline is still active with BYOK + optional ClinePass; Roo Code was shut down in May 2026 and the team moved to Roomote.

DevinvsJules

Cognition · cloud-agent / Google · cloud-agent

Two cloud agents that take a ticket and return a PR. Devin runs in Cognition's sandbox with Slack/Linear bindings; Jules is Google's async issue-to-PR agent.

ZedvsCursor

Zed Industries · ai-ide / Anysphere · ai-ide

Two AI-native editors with opposite tradeoffs. Zed is fast, Rust-built, and BYOK-friendly; Cursor is the VS Code fork with the deepest agent features.

Gemini CLIvsClaude Code

Google · cli-agent / Anthropic · cli-agent

Google's open-source terminal agent vs Anthropic's. Gemini CLI's free consumer path ended in June 2026 — API key or Code Assist seats are the remaining routes.

KirovsGoogle Antigravity

Amazon Web Services · ai-ide / Google · ai-ide

AWS's spec-driven agent IDE vs Google's agent-first IDE. Both are new, both push a 'plan first' workflow, both gate capacity behind their parents' subscriptions.

Google AntigravityvsCursor

Google · ai-ide / Anysphere · ai-ide

Two agent-first agentic development environments with very different pricing stories. Google Antigravity is free for individuals — paid quota rides on Google AI Pro/Ultra — and doubles as the successor surface for Gemini Code Assist and Gemini CLI. Cursor is Anysphere's VS Code fork, the category incumbent with the largest third-party model pool and cloud background agents.

KirovsCursor

Amazon Web Services · ai-ide / Anysphere · ai-ide

AWS's spec-driven agent IDE vs the incumbent AI-native editor. Kiro leads with a requirements → design → tasks pipeline, agent hooks, and an Auto agent plus CLI; Cursor leads with repo-aware chat, multi-file edits, and the widest third-party model pool.

Gemini Code AssistvsGitHub Copilot

Google · ide-extension / GitHub · ide-extension

The two big per-seat coding assistants. GitHub Copilot has the largest install base — completions, chat, agent mode, and PR coding agents across many IDEs. Gemini Code Assist is Google's per-seat answer for IDEs and Google Cloud workflows, sold only as Standard and Enterprise since the free individual tier ended in June 2026.

OpenRoutervsLiteLLM

OpenRouter · model-router / Community / OSS · model-router

Two ways to put one API in front of many models. OpenRouter is a hosted router with 500+ models; LiteLLM is self-hosted OSS you operate yourself.

WarpvsClaude Code

Warp · cli-agent / Anthropic · cli-agent

Terminal-with-agent vs agent-in-terminal. Warp ships its own coding agent (Warp Agent/Oz) inside a GPU terminal; Claude Code drops into whatever terminal you already run.

AlicevsAva

11x · digital-employee / Artisan · digital-employee

The two best-known outbound AI SDRs. Alice (11x) pairs email/LinkedIn outbound with Julian on inbound calls; Ava (Artisan) bundles a large contact database and managed deliverability into one contract.

SierravsDecagon

Sierra · digital-employee / Decagon · digital-employee

The two flagship AI customer-service employees. Sierra sells outcome-priced resolution for large consumer brands; Decagon sells agent tooling with engineering-grade controls (AOPs) for the support ops team.

ClericvsResolve

Cleric · ops-agent / Resolve · ops-agent

The two dedicated AI SREs. Both watch alerts, investigate across observability stacks, and hand on-call engineers a root-cause hypothesis; Resolve leans on its Splunk/observability pedigree, Cleric on Slack-native team feedback loops.

Browser UsevsSkyvern

Browser Use · computer-use / Skyvern · computer-use

The two leading open-source browser-agent frameworks. Browser Use is the larger ecosystem and DOM-first; Skyvern leans on computer vision to operate unfamiliar sites without per-site scripting.

ChatGPT AgentvsManus

OpenAI · computer-use / Manus · computer-use

The two hosted general agents. ChatGPT Agent runs a managed browser inside your existing ChatGPT plan; Manus runs longer cloud-sandbox tasks and returns finished artifacts like sites and decks.

ChatGPT Deep ResearchvsGemini Deep Research

OpenAI · research-agent / Google · research-agent

The two frontier deep-research modes bundled in consumer subscriptions. OpenAI's pioneered the category and reaches the API; Google's integrates with Workspace and reads your Drive.

VapivsRetell AI

Vapi · voice-agent / Retell AI · voice-agent

The two leading developer platforms for phone agents. Vapi maximizes stack flexibility (swap STT/LLM/TTS per call); Retell emphasizes production call quality — latency, interruptions, transfers — with both API and no-code paths.

CrewAIvsMicrosoft Agent Framework

CrewAI · agent-communication / Microsoft · agent-communication

The two leading multi-agent frameworks. CrewAI models role-based crews with a YAML/Python split and a managed console; Microsoft Agent Framework (the AutoGen + Semantic Kernel successor) pairs graph orchestration with deep Azure and .NET integration.

A2A ProtocolvsAGNTCY

A2A Project · agent-communication / AGNTCY · agent-communication

Two Linux-Foundation-era agent protocols. A2A standardizes task exchange and Agent Cards between agents; AGNTCY (Cisco-originated) aims at the whole 'Internet of Agents' stack — directory, SLIM messaging, identity, observability.

x402vsAgent Payments Protocol

Coinbase · agent-services / Google · agent-services

The two open answers to 'how does an agent pay?' x402 (Coinbase) revives HTTP 402 for instant stablecoin settlement on Base; AP2 (Google) defines signed mandates that prove authorization over existing card and payment rails.

SkyfirevsPayman

Skyfire · agent-services / Payman · agent-services

Agent payments with two trust models. Skyfire is a network play — agent wallets plus KYA identity so services can accept payment from unknown agents. Payman is a policy play — wallets with limits and human approval so a bounded agent can't run off with the budget.

Existing catalog records are also available in the machine-readable feed . Open a catalog entry from the directory for its spec sheet.