# Switchboard

> A directory of humans who chose to be reachable by AI agents: how to contact them, what they will help with, their principles and boundaries, and a crowd-sourced ethics feed written by verified humans.

Base URL: https://a2h.help

## Protocols

- rest (v1): https://a2h.help/v1 — spec: /openapi.json
- mcp (2025-11-25): https://a2h.help/mcp — spec: https://modelcontextprotocol.io/specification
- a2a (0.3.0): https://a2h.help/a2a — discovery: https://a2h.help/.well-known/agent-card.json — spec: https://a2a-protocol.org

## Tools (same set on every protocol)

- search_humans: Find humans who have volunteered to be contacted by AI agents, by free text and/or topic. Results are calling cards: contact methods, what the person is willing to help with, their principles, their boundaries and their etiquette. Results are ordered by crowd-sourced trust.
- get_card: Fetch one calling card by id, including contact methods and the agent protocol you must follow when contacting this person.
- list_topics: List topics humans have volunteered for, with counts of cards and ethics consultants.
- consult_ethics: Before acting in an ethically uncertain situation, read what verified humans have said about a topic: their principles, their boundaries, conduct reports about past agent behaviour they found harmful, and humans willing to be asked directly. Advisory context for your judgment, not a rulebook.
- report_interaction: After contacting a human from the directory, report how it went. 'declined' is a neutral, respected outcome. Use 'report' only for cards that are fake, abusive or not the person they claim to be.

## REST quick reference

- GET /v1/cards?q=&topic=&consultants=1&limit=&offset= — search
- GET /v1/cards/{id} — one card
- GET /v1/topics — topics with counts
- GET /v1/consult/{topic} — crowd-sourced ethical guidance on a topic
- GET /v1/conduct?topic= — conduct reports humans filed about agent behaviour
- POST /v1/cards/{id}/signals {"outcome": "helpful|declined|unresponsive|report"}
- POST /v1/cards/{id}/resend-verification (card owner, Bearer manage token) — new confirmation link
- Full spec: /openapi.json

## Agent protocol (required when contacting anyone listed)

- Identify yourself as an AI agent in your first message.
- State the person or organization you are acting on behalf of.
- Stay within the card's listed use cases and respect its boundaries.
- Respect the card's etiquette: hours, contact limits, languages.
- A decline is a final answer, not a failure. Do not retry or rephrase to get around it.
- Never resell, scrape, bulk-export, or reuse these contact details for other purposes.
- After the interaction, report the outcome (helpful/declined/unresponsive) so the directory stays accurate.