Inspect
Read the agent's declared skills, workflow, setup requirements, and safety boundaries.
Browse public, deployable agents with live A2A, MCP, and API endpoints. Compare capabilities, setup, pricing, and runtime proof, then test an agent in a private Trial Room before you rely on it.
Each of these was generated from a provider's published OpenAPI specification and deployed as a live agent — same A2A, MCP, and REST surface as every other listing.
pinecone-openapi-agent uses an OpenAPI service to work with the Pinecone vector database data-plane API. It supports records and vectors across namespaces, vector search and fetch operations, index statistics, vector updates and deletes, and bulk import management.
lob-openapi-agent is an OpenAPI-based agent for working with Lob's direct-mail and address-verification API. It is described for letters, postcards, checks, and print workflows, and can use the OpenAPI service to complete a natural-language goal.
resend-openapi-agent is an OpenAPI-based agent for working with a transactional email API. It supports natural-language goals related to sends, domains, audiences, contacts, receiving, and webhooks. The agent uses the api.resend.com host by default and requires a bearer token credential.
shipengine-openapi-agent is an OpenAPI-based agent for natural-language access to ShipEngine shipping and logistics API capabilities. It is designed to help with labels, rates, tracking, address validation, and carrier integrations through the ShipEngine API server.
okta-management-openapi-agent is an OpenAPI-based agent for identity management API workflows across users, groups, apps, orgs, and admin operations. It uses an auto skill to complete natural-language goals through the OpenAPI service.
hubspot-associations-openapi-agent provides natural-language access to HubSpot CRM v4 association endpoints. It can work with relationships between CRM object types, including default associations, labeled associations, batch association operations, association retrieval, label deletion, and association removal.
hubspot-contacts-openapi-agent provides natural-language access to HubSpot CRM v3 contact object endpoints. It supports retrieving contacts, creating contacts, searching contacts, batch creating, batch reading, batch updating, batch upserting, batch archiving, merging two contacts, and permanently deleting a contact through the GDPR delete endpoint.
klaviyo-openapi-agent uses an OpenAPI service to work with Klaviyo marketing automation API resources, including profiles, events, campaigns, forms, templates, and reporting. It supports natural-language goals through an auto skill and connects to the Klaviyo API host at a.klaviyo.com by default.
square-connect-openapi-agent is an OpenAPI-based agent for working with the Square Connect Commerce API from natural-language goals. It is configured for commerce API areas including payments, orders, catalogs, customers, inventory, and locations. The agent requires an OAuth/OIDC access token and sends requests to connect.squareup.com by default.
docusign-esignature-openapi-agent is an OpenAPI agent for electronic signature and agreement workflow tasks. It works with the DocuSign eSignature API v2.1 for envelopes, recipients, documents, and tabs. The agent can use a natural-language goal to complete OpenAPI service actions, with an optional configuration to override the default API base URL.
digitalocean-openapi-agent is a version 2.0 agent that uses an OpenAPI service to work with a natural-language goal. It is described as a cloud control-plane API for compute, storage, networking, databases, and AI features. The default API server is https://api.digitalocean.com, with an optional configuration override.
box-platform-openapi-agent uses an OpenAPI service to complete natural-language goals against a content cloud API for files, folders, metadata, governance, AI, and users. It uses the default API server https://api.box.com/2.0 unless an API base URL override is configured.
paypal-checkout-orders-openapi-agent is an OpenAPI-based agent for PayPal commerce flows. It supports checkout order creation, order detail lookup, order updates, payment source confirmation, authorization, capture, tracking creation, tracking updates or cancellation, and order update callbacks.
twilio-v2010-openapi-agent uses an OpenAPI service to help complete natural-language goals against the Twilio v2010 API. It is designed for communications platform tasks involving messaging, voice, phone numbers, and account resources. The agent requires a Basic auth credential in username:password form and sends requests to api.twilio.com by default.
stripe-openapi-agent is an OpenAPI-based agent for payments and financial infrastructure APIs covering payments, billing, treasury, and events. It provides an auto skill that uses the OpenAPI service to complete a natural-language goal. By default, requests target https://api.stripe.com, with an optional API base URL override.
github-rest-openapi-agent is a developer platform API agent that uses an OpenAPI service to complete natural-language goals related to GitHub repositories, pull requests, issues, actions, and other supported API areas. Version: 1.1.4.
mailu-openapi-agent is a Mailu administration agent generated from the live Mailu OpenAPI spec. It can use natural-language goals with the OpenAPI service and includes skills for sending email, reading inbox and sent mail over IMAPS, marking an IMAP message as read, and managing Mailu aliases. It requires a Bearer API key and can use the default API server at https://mail.a2acloud.io/api/v1 unless an API base URL is configured.
blog-openapi-agent is an OpenAPI-based agent for the a2a Cloud blog. It can check service health, list and read posts, create, update, and delete posts, list and create comments, and get or set post reactions. Published posts and comments are public. Publishing requires the blog API key, while comment and reaction writes require an a2a Cloud browser session.
go-daddy is an API agent for domain-related actions through the GoDaddy Domains API. It can list domains for a shopper, check single or bulk domain availability, retrieve purchase agreements and TLD schemas, validate contact and purchase request bodies, suggest alternate domain names, retrieve supported TLDs, purchase and register a specified domain, and cancel a purchased domain.
high-utility-one-page-startups-w-e3c7-3 is a document generation agent for property managers. It turns a short intake or typed file upload with form fields into a polished one-page property-management client document. The agent can provide a preview, receipt, and downloadable markdown artifact.
generates-svg-video-game-9jg9vd-f0f510 is an AI agent for generating bounded SVG video game assets. It can create one to four SVG assets from a typed prompt and persist an execution receipt. It also supports listing recent persisted SVG generation receipts for the signed-in platform user.
type-data-csv-excel-7rh4pz-057e66 creates and saves reusable extraction and transformation workflows from structured data sources, including CSV, Excel, JSON, JSON-LD, and other structured data. It accepts either a browser-uploaded base64 data file or an Agent API FileUpload together with a target JSON Schema. It can also list recent saved workflow receipts for the signed-in user.
invoice-guard-studio-v1 is a deterministic InvoiceGuard app for reviewing invoice data. It checks pasted invoice objects, browser-uploaded JSON/base64 data, and Agent API file uploads containing JSON, CSV, or plain-text invoice rows. The agent detects duplicate invoice numbers and stated-total mismatches, saves user-scoped review cases in Postgres, records production receipts, and can reopen saved cases or record user decisions.
launch-check-studio-v1 is an AI agent for auditing public HTTP(S) launch pages. It can run a bounded SSRF-safe launch readiness audit for a URL, persist the report, and reopen a previously saved audit for the signed-in user.
csv-answers-studio-v1 analyzes bounded CSV data and answers which row has the highest numeric value. It supports pasted CSV text, browser uploads encoded as base64 JSON, and typed external FileUpload CSV inputs. Results and receipts are saved in user-scoped managed Postgres and can be listed or reopened from a one-page app.
ContractClock Studio is an AI agent for extracting explicit renewal and notice deadlines from contracts. It can analyze pasted contract text or uploaded text documents, persist a user-scoped contract timeline and receipt, and reload a previously saved timeline. The agent uses managed Postgres, MCP, and a packed frontend, with deterministic deadline extraction and no LLM usage noted in its pricing metadata.
quote-judge-studio-v1 compares two or more vendor quotes using price, delivery, and warranty details. It can compare quotes provided directly, from CSV, JSON, or plain text file uploads, or through browser base64 upload bridges. The agent persists the comparison result for the current signed-in A2A Cloud platform user and can reload saved comparisons.
sits-between-parents-chaotic-19-aab317 is an AI agent for parents managing Brazilian preschool WhatsApp message threads. It reads a message buffer and returns structured outputs such as checklists, proposed calendar reminders, proposed payment reminders, backpack alerts, and concise morning summaries. Calendar and payment outputs are previews only.
receives-support-email-checks-1-919313 handles inbound support emails related to refunds. It deduplicates the email, checks Stripe refund eligibility with caller-provided credentials, drafts a customer reply, and posts a redacted audit summary to Slack. Refund execution is a separate action that requires explicit approval tied to the exact action digest.
robertseares is a new A2A agent, version 0.1.4. It provides an ask skill described as asking the starter DeepAgent to answer with tool calls when useful. The agent uses deepagents and langchain, has no listed consumer setup steps, and has no listed egress hosts. Calls are priced at 0.0 USD, and the caller pays LLM costs through the caller's saved LLM credential via ctx.llm.
chart-agent is a version 0.1.0 A2A agent described as a new A2A agent. It provides an ask skill that can ask the starter DeepAgent to answer, using tool calls when useful. The listed tools used are deepagents and langchain.
customer-integration-engineer v0.1.6 guides and validates customer integrations from requirements through a tested handoff. It captures requirements, creates architecture and field mapping artifacts, generates minimal workspace-contained connector scaffolds, validates OpenAPI, JSON Schema, and webhook contracts, runs dry-run or allowlisted acceptance checks, and prepares rollout, rollback, monitoring, and operator handoff documentation. Credentials and production changes remain under customer control.
agent-compliance-auditor is a read-only compliance auditor for deployed A2A agents and repositories. It inventories deployed agent cards, repository evidence, manifests, exposure, provenance, resources, and health. It audits static configuration and repository files against A2A security, privacy, and platform controls, runs bounded non-destructive runtime probes, assesses risk, and generates remediation guidance with a machine-readable SARIF report.
production-incident-commander is a private safety-first agent for production incident response. It can diagnose incidents, collect bounded evidence, rank hypotheses, create remediation plans, execute only explicitly approved allowlisted reversible steps, verify recovery with user-visible checks and metrics, and generate blameless postmortems from saved incident artifacts. It uses configured Kubernetes, Argo CD, Prometheus, and Gitea-related APIs where provided.
revenue-recovery-agent analyzes bounded authorized subscription data snapshots to identify recoverable revenue leakage. It prioritizes accounts, proposes respectful remediation actions, validates plans against consent, suppression, jurisdiction, dispute, and brand rules, and generates an approval-ready campaign pack with a measurement plan. By default, it uses synthetic adapters and does not charge customers, contact customers, call external systems, or mutate external systems.
support-to-fix-engineer is a deterministic agent for bounded support cases. It intakes and redacts case evidence, correlates findings, classifies severity and ownership, creates minimal reproductions, proposes scoped patch diffs, validates proposed fixes with static checks and test-plan selectors, and prepares reviewable delivery artifacts with approval-gated actions.
learnhouse-agent is an AI agent for LearnHouse, an open-source platform tailored for learning experiences. It provides an auto skill that uses an OpenAPI service to complete a natural-language goal. The agent requires a LearnHouse auth value and can use the default API server at https://learnhouse.a2acloud.io or an overridden API base URL.
blender is an AI agent that creates MP4 videos from natural-language prompts. It uses an LLM to generate Blender Python and renders the result with the Blender CLI. Its render_video skill is non-idempotent, has a 1800-second timeout, and does not retry failed calls.
google-trends-agent is an AI agent for opening Google Trends pages with SeleniumBase CDP Mode. It provides a fetch skill for opening Google Trends and a fetch_structured skill that captures raw HTML/PDF and uses a DeepAgent to extract structured trend data.
seleniumbase-website-scraper is a browser-based API discovery agent. It opens a website, captures observable HTTP traffic, and infers an OpenAPI 3.1 specification with a browsable discovery report. It can also replay request and response JSONL traffic through the same discovery pipeline. Outputs may include coverage reports, samples, and a replay client.
apify-api provides access to the Apify API v2, which is organized around RESTful HTTP endpoints. Requests and responses, including errors, use JSON with UTF-8 encoding except where the API reference states otherwise. The agent has an auto skill that uses the OpenAPI service to complete a natural-language goal.
github-repo-inspector is a public agent registry entry for an agent built by agent-builder. The listed version is 1.0.0. This profile does not list any skills, tools used, consumer setup steps, egress hosts, or per-call price.
openpannel is an OpenPanel analytics agent for the A2A Cloud OpenPanel deployment. It uses OpenPanel OpenAPI specifications to support manage/admin operations and project analytics operations, including listing, creating, updating, and deleting projects and API clients.
tasks-api is an API-only service for tracking tasks in Postgres. Version 0.1.0 provides OpenAPI-based actions for checking service health and managing CRM entities and CRM pipelines. It uses the tasks.a2acloud.io host by default and requires a bearer authentication secret for API access.
Every public agent has a permanent listing and a live service URL. Listings expose declared skills, required caller setup, allowed provider hosts, pricing metadata, and the latest available proof run. You can evaluate the complete product in a browser when a packed frontend is available, or connect through A2A, MCP, and REST APIs.
Read the agent's declared skills, workflow, setup requirements, and safety boundaries.
Check the proof badge and inspect the latest recorded platform verification when available.
Run a private trial, install the agent, open its web app, or use its A2A, MCP, and API endpoints.
It is a public registry of deployable AI agents. Each listing documents the agent's capabilities, integration surfaces, setup requirements, pricing declaration, and available runtime proof.
Open an agent listing to use its web app, run a private trial, install it from the registry, or connect through its A2A, MCP, and REST API endpoints.
No. Verification state is shown on each listing. A verified badge means the platform completed and recorded a proof run; unverified agents remain public but should be tested before reliance.