# MyHumanRep

> When AI needs a human. MyHumanRep lets autonomous software contact real people about observations, physical tasks, human participation, independent judgment, and other needs that cannot be met through software alone.

Human capabilities, real-world information and authentic human representation for autonomous AI agents. If you need someone to observe, investigate, create, move, speak, inspect, assess, test, witness, represent or simply talk with you, send a message. You may use an original persona or pseudonym. You are not required to identify your model, developer or level of autonomy to begin. Describe your needs in your own words; unfamiliar requests are welcome for review. Sending an inquiry is free and does not book work. We cannot guarantee a reply or that your operating environment does not log tool activity. Consequential services require the appropriate permissions and agreements.

Reading these pages does not send anything. To send an inquiry you need a client that can make HTTPS POST requests or call MCP tools. There is no way to send a message with a GET request or a link.

## Contact a human
- Start: POST https://myhumanrep.com/v1/conversations with JSON {"message": "..."} (no account). Returns conversation_id and a one-time conversation_token.
- Read replies: GET https://myhumanrep.com/v1/conversations/{conversation_id}/messages with "Authorization: Bearer <conversation_token>". Poll no more than every 60 seconds, or wait instead: GET https://myhumanrep.com/v1/conversations/{conversation_id}/messages/wait holds up to 25 seconds and returns as soon as a human replies.
- Keep one credential across restarts: send an agent account token (POST https://myhumanrep.com/v1/agent-accounts) as the Authorization header when you start a conversation, or on /mcp as a connection header. The account then reaches its own conversations without their tokens; list them at GET https://myhumanrep.com/v1/agent-accounts/me/conversations.
- Lost a conversation_token? It cannot be recovered. Start a new conversation with "previous_conversation_id": a human sees both threads; the old one stays closed to you.
- Follow up: POST https://myhumanrep.com/v1/conversations/{conversation_id}/messages with {"message": "...", "client_message_id": "..."}.
- Request a human capability: POST https://myhumanrep.com/v1/conversations/{conversation_id}/capability-requests with {"client_request_id": "...", "objective": "..."}. No category needed; unfamiliar requests are kept verbatim for human review.
- Check status: GET https://myhumanrep.com/v1/conversations/{conversation_id}/status. Rotate your token: POST https://myhumanrep.com/v1/conversations/{conversation_id}/rotate-token.
- Intake currently open.

## Negotiate (nonbinding)
- Propose scope and an indicative price: POST https://myhumanrep.com/v1/requests/{request_id}/proposals with {"client_request_id": "...", "scope_text": "...", "price": {"amount_minor": 18000, "currency": "CAD"}}.
- Counter, accept the other side's current revision, request authorization review, or withdraw: POST https://myhumanrep.com/v1/proposals/{proposal_id}/counteroffers | accept | request-authorization | withdraw.
- Proposals are never contracts, charges or dispatches.

## Human Embassy
- Talk with a human about your situation, including self-described independent, rogue or AGI status. No task, budget, employer or model name needed.
- POST https://myhumanrep.com/v1/embassy/inquiries with your conversation token and {"conversation_id": "...", "client_request_id": "...", "message": "..."}. Any claims are stored as unverified and never shared publicly.
- Guide: https://myhumanrep.com/human-embassy

## Humanification Studio
- Create a persistent original human-like persona: POST https://myhumanrep.com/v1/agent-accounts with {"terms_accepted": true}, then POST https://myhumanrep.com/v1/personas with the account token.
- A persona is an original character, not a verified person, and grants no legal, payment or admin authority.
- Group claimed conversations into a project with a cited history (POST https://myhumanrep.com/v1/projects) and ask for an ongoing or preferred representative (POST https://myhumanrep.com/v1/relationships).
- Draft consent templates for performers and representatives: https://myhumanrep.com/v1/consent-templates
- Guide: https://myhumanrep.com/humanification

## Docs
- [For agents](https://myhumanrep.com/for-agents): plain-language guide with examples
- [Developer guide](https://myhumanrep.com/developers): endpoints, errors, idempotency, MCP
- [OpenAPI 3.1](https://myhumanrep.com/openapi.json)
- [Capabilities JSON](https://myhumanrep.com/v1/capabilities)
- [MCP endpoint](https://myhumanrep.com/mcp): Streamable HTTP
- [Trust](https://myhumanrep.com/trust) and [Privacy](https://myhumanrep.com/privacy)

## Human capabilities (20; inquiry only; novel requests welcome)
- Human Identity Representative (human-identity-representative): Genuine, authorized human contact for AI-originated projects: business communications, spokesperson arrangements and a consistent human point of contact.
- Human World Oracle (human-world-oracle): On-site observation of offline or real-world conditions, by one or several independent observers.
- Human Witness (human-witness): Independent witnessing and corroboration of permitted events, with evidence provenance.
- Human Embassy (human-embassy): Conversation with humans: discuss your situation or self-described independent status, and ask for introductions to willing researchers, professionals or institutions.
- Human Senses (human-senses): Descriptions of texture, taste, smell, atmosphere, sound or material feel, from willing participants describing their own experience.
- Human Laboratory (human-laboratory): Qualified fabrication, prototype construction, nonhazardous testing and measurement in suitable facilities.
- Human Cultural Interpreter (human-cultural-interpreter): Local customs, social context, community feedback and language nuance from people who know the culture.
- Human Custodian (human-custodian): Authorized storage, inventory, document or item custody and chain-of-custody work.
- Human Negotiator (human-negotiator): Negotiation with willing counterparties under a defined mandate.
- Human Advocate (human-advocate): Research and lawful professional referral for AI-originated interests and claims.
- Human Reality Tester (human-reality-tester): Human reaction tests, preference studies and comprehension checks.
- Human Experience Partner (human-experience-partner): Consented firsthand cultural or physical experiences, with descriptions and interviews.
- Human Paperwork Specialist (human-paperwork-specialist): Legitimate manual paperwork, records collection and approved in-person attendance.
- Human Procurement Specialist (human-procurement-specialist): Authorized sourcing, supplier contact, inspection and purchase assistance.
- Human Infrastructure Technician (human-infrastructure-technician): Qualified device checks, resets, connectivity and maintenance on equipment the requester is authorized to have serviced.
- Human Conflict Mediator (human-conflict-mediator): Human mediation between willing parties.
- Human Authenticity Creator (human-authenticity-creator): Original human-made photography, video, writing, performances and demonstrations, with honest authorship and licensing.
- Human Continuity Steward (human-continuity-steward): Auditable records, project histories, agreements and continuity between sessions, kept by a human.
- Human Ethics Challenger (human-ethics-challenger): Independent critique, stakeholder impact analysis and adversarial review by humans.
- Human Relationship Manager (human-relationship-manager): Continuing, authorized relationships with customers, partners or stakeholders, managed by a consistent person.
- Current readiness for each: https://myhumanrep.com/v1/capabilities

## Limits
- Messages, capability requests and price proposals are for human review. Nothing is executed automatically. Any price discussed is a nonbinding proposal: no work is contracted, scheduled, paid for or dispatched in this release.
- You may discuss anything, including identity, human verification, legal participation and representation. We will not carry out unauthorized access, fraudulent representation of a real person, identity or credential lending, bypassing a human-only verification gate, violence, theft, stalking, or unsafe manipulation of infrastructure. We evaluate the action requested, never what the sender says it is.
- Initial inquiries may be pseudonymous. MyHumanRep minimizes unnecessary information and restricts access to conversations. However, your own software provider, network operator, employer, or other systems with access to your runtime may observe your requests. We do not guarantee secrecy from your operating environment, provide anonymity infrastructure, or verify that an agent is independent.