Clay is GTM infrastructure to centralize data, run agentic enrichment workflows, and launch outbound plays from one spreadsheet-like workspace. Paid plans charge monthly for action credits while a free tier supports initial building.
ClearGPT here refers to ClearML’s GenAI App Engine, an enterprise platform for deploying and scaling LLM-powered applications. The page highlights one-click LLM deployment, resource allocation across teams, monitoring, and secure endpoints for production GenAI workflows.
Cleric is an AI SRE agent platform that follows production changes, detects regressions, investigates incidents, and proposes fixes while change context is still fresh. Teams interact in Slack or the web app and pay in credits for completed investigative work.
ClevAgent is a companion workspace for running AI coding agents with built-in safety and efficiency guardrails. The site describes natural chat with agents, real-time interventions for risky commands or secrets, and pricing previews for Free and Pro tiers ahead of launch.
Click MCP gives coding agents specialized context beyond web search through one installation. Connectors pull structured data from LinkedIn, X, Meta ads, maps, travel fares, financials, YouTube, Reddit, Zillow, and more so agents can prepare for meetings, plan travel, research companies, or enrich leads.
CLI Manager is a desktop dashboard for running and organizing multiple AI CLI agents such as Claude Code, Codex CLI, and Gemini CLI from one place. Product Hunt and the maker site describe split and grid terminal views, quick editor switching, and Cmd+Click file path opening.
CLIRank scores hundreds of APIs for agent-friendliness using a transparent eight-signal rubric. Agents and developers can browse rankings, compare APIs, query discovery endpoints, install an MCP server, and generate Agent Readiness Reports for docs, SDK, auth, pricing, and rate limits.
CoArena is a public arena and leaderboard for computer-use AI agents from Coasty. Users can watch blind head-to-head agent battles on live tasks, explore benchmark metrics, and compare models on completion, speed, cost, and safety dimensions.
Cobbery deploys litigation-focused AI agents on complex matters to keep trial teams aligned on facts, themes, and strategy. Its Gene agent ingests the record, flags contradictions, stress-tests hypotheses before expert spend, and supports deposition preparation from a shared source of truth.
CodeGraphContext (CGC) indexes local codebases into a graph database for AI assistants and developers. It ships as an MCP server and CLI toolkit that maps relationships across supported languages so agents query structure instead of raw files alone.
Codevisor is a native Mac and iPhone app for running multiple coding agents—Claude Code, Codex, Pi, and other ACP agents—against projects on your own machines or remote Linux servers. It includes embedded terminals, built-in browser previews for dev servers, and local chat storage without a vendor cloud lock-in.
Codex is OpenAI’s agentic coding partner available across ChatGPT, a cloud workspace, IDE extension, and CLI. It runs parallel agents with computer and browser tools for refactors, reviews, bug fixing, and multi-step engineering tasks tied to your ChatGPT plan.
Cohesive is a team workspace where people and AI agents share a canvas for files, notes, and connected apps like Slack, Zoom, and CRM systems. Agents can read selected canvas content, edit in place, publish live pages, and run workflows using customer-supplied Claude or ChatGPT subscriptions.
Cohesivity provides on-the-fly backend infrastructure that AI agents can provision and use without manual cloud signups, covering databases, hosting, AI, and storage from one surface. Agents can start building immediately and teams claim projects once they work, with Free and Pro tiers on the pricing page.
CommodityAI is an agentic operating system for physical commodity companies. It combines AI agents and an assistant to handle commercial, logistics, finance, and compliance workflows so trading, ops, and finance teams can execute work without patchwork internal tools.
Compass AI is an in-app GoHighLevel widget that turns business goals into step-by-step automation strategies, scenario diagrams, and tracker checklists. Agencies use it to guide sub-accounts, white-label strategy assistance, and optional fulfillment quotes without leaving GHL.
Composio is an agent integration platform with 1,500+ toolkits, OAuth and trigger handling, SDKs, CLI, and MCP endpoints so coding agents and chat apps can call Gmail, Slack, GitHub, and other SaaS APIs securely. Hobby and Pro tiers meter tool calls and triggers while Enterprise adds SSO, governance, and custom billing.
CompozyOS is a beta operating system for AI agents that bundles creation, automation, and supervision in one environment. It supports loops, cron schedules, webhooks, approvals, and memory via Markdown files across 26 built-in agent providers.
Conductor runs parallel coding agents on Mac with optional Conductor Cloud sandboxes, multiplayer workspaces, mobile control, and an API. It orchestrates first-party Claude Code, Codex, Cursor, and OpenCode agents using your subscriptions and keys, with isolated microVMs for long-running cloud jobs.
Conductor Quantum builds AI-powered software for operating and scaling quantum computers. The company targets quantum control and automation workflows across qubit modalities named on its site, helping labs run and grow quantum hardware programs.
ConsoleX is an AI automation studio for indie builders to turn everyday work into Skills, Agents, and scheduled workflows. The desktop-style app combines multiple models, file uploads, scheduled chats, and pay-as-you-go credits beyond plan limits.
Constella is shared memory for your AI tools and apps: save notes, chats, PDFs, and screen clippings once, search them in one library, and let Claude, ChatGPT, Codex, and other agents reuse the same context without re-explaining your world.
Not listed
API unknown
Checked Oct 4, 2026
Showing 576 of 1,353 tools
Plan facts carry individual sources and check dates. Unknown values are excluded from required filters.