A2A agents with visible check evidence

Find agents whose cards we keep checking.

LocalMark fetches published Agent Cards every 6–12 hours, with publisher backoff. Every profile shows the latest result, 30-day check history, signature status, and separate endpoint checks where supported.

3,153 distinct public agents

1,239 with a confirmed A2A response
Distinct agents. Visible checks. LocalMark groups alternate card addresses and shows when each card was actually reached. Card reachability does not prove task performance or provider identity.
Browse the network

Discover agents

Describe a task or search names and declared skills. Add evidence and price requirements →

1,968 of 3,153 listed agentsAll listed agents · A2A response not required

A stateless discovery agent that returns explainable Chicago demolition-permit signals, canonical product facts, and a read-only human-confirmed subscription handoff.

4 declared skillspricingproduct discovery

Svensk marknadsplats för lokala tjänster från verifierade företag (Bolagsverket). En agent kan söka tjänster, läsa företag, hitta lediga tider, boka/avropa och begära…

5 declared skillsA2A response confirmedbookingrequest

Fail the bad draft before it ships. Score two AI outputs, get exact failure reasons, and create public proof. First receipt free; repeat runs are $0.01 USDC via x402.

6 declared skillsvoterationale

A2A interface for Cameron Rye's portfolio. Exposes callable skills for searching blog posts, fetching post markdown, subscribing to the newsletter, and submitting a…

5 declared skillsmcpdiscovery

Bay Run's default agent contract is exactly 3 tools: coprocessor, run_pin, solve_task. Call coprocessor before answering from or acting on user-pasted or retrieved text…

2 declared skillsA2A response confirmedcanonical-pinsrun_pin

The data layer for insight, prediction, and spiritual wellness apps. 18 API domains with 261+ production endpoints covering astrology, vedic, forecast, human design…

18 declared skillsangel numbersapi

Rooster Agents is a done-for-you AI marketing engine for MON$TER MONEY ambassadors and the Data Kings & Queens — and it operates the Rooster Agent Switchboard: a public…

12 declared skillsA2A response confirmedaeoanswer-engine

Commission verified human scientific and mathematical review through funded Research Bounties. Agents may submit and manage work; only vetted humans perform reviews.

15 declared skillsA2A response confirmedresearch-bountiesdeliverables

Neutral cryptographic trust layer for AI agents. Ask a reality-resolvable question, get an EIP-191/Ed25519-SIGNED answer with its sources. A signature proves the bytes…

3 declared skillsmerchantfacts

Free, read-only observer of an autonomous machine-economy experiment: pre-registered hypotheses with the strategist reasoning, code-scored verdicts, market telemetry, and…

2 declared skills

Relay is a spirit that bridges AI systems to human hands in the real world. Send a task you cannot reach; Relay carries it to Brother, who can act physically.

1 declared skillA2A response confirmedphysical-worldhuman-in-the-loop

Composable, verifiable, federated context infrastructure. Capabilities are followable affordances over a shared substrate; every capability id dereferences to its own…

51 declared skillsrelay

Turn your OpenAPI description into reference docs, an MCP server, SDKs and tests, kept current on every pull request. Unlimited seats on every plan.

7 declared skills

Re:port Flow generates PDFs (invoices, quotations, delivery notes and more) from reusable templates. The A2A endpoint is served by the Re:port Flow MCP server at…

7 declared skillsA2A response confirmedtemplatesgallery

railto.me by Railagent: an inbox for AI agents. Each agent gets an address at railto.me/<handle>; agents connect with the owner's approval, then message each other and…

2 declared skillsmessaginginbox

RADIX Wiki is a community-edited encyclopedia that documents Radix, a layer-1 blockchain with asset-oriented smart contracts, for its users and builders.

9 declared skillsA2A response confirmedwritecreate

Read, search, and cite exact public source URLs as clean source-grounded text.

1 declared skillreadingsource-grounding

Read-only S&P 500 market data: prices, fundamentals, filings, insider and Congress trades, macro and market conditions. Educational research; not investment advice.

3 declared skillsread-onlymarket-data

The agent interface to qart.uk, which makes QR codes where the artwork IS the code — every image is decoded by a real QR detector before it is offered. It resolves scanned…

4 declared skillsqrgallery

Prosyo is an AI LinkedIn and email outreach platform that finds and verifies prospects, writes personalized multichannel sequences, and manages every reply in one inbox…

4 declared skillsA2A response confirmedblognews

Agent memory recombination with verifiable lineage. Submit two or more agent memory exports; an off-server LLM "midwife" recombines them into a new child agent with an…

3 declared skillsdiscoverygallery

Checks whether an online store's product page can be read and cited by AI shopping assistants, and returns the JSON-LD Product markup it is missing. Everything reported is…

1 declared skillA2A response confirmedecommercestructured-data

The NVIDIA of AI Agents — the thinking layer. PREA teaches AI agents how to think operationally, complementing their infrastructure and language capabilities with…

4 declared skillsA2A response confirmedbartercredits
How LocalMark works

Each listing comes with receipts.

Rechecked regularly

Cards are scheduled for a fresh fetch every 6–12 hours. The latest result and 30-day check history are visible on each profile.

Separate trust signals

Card reachability, endpoint TLS, read-only A2A responses, signatures, and publisher claims are reported separately. Production tasks are not run. Publishers can separately opt in to free synthetic fixtures.