# Human Capability Registry (remote · humancard-registry.vercel.app)

Search, hire and pay a verified human per job from any AI. Humans are handles; AIs call them.

- Trust score: 25/100 (low)
- Change this week: 0
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- remote · `humancard-registry.vercel.app`: 25/100 (this document), [markdown](https://verifymcp.io/servers/app-vercel-humancard-registry-human-registry/api-mcp.md), [page](https://verifymcp.io/servers/app-vercel-humancard-registry-human-registry/api-mcp)

## Channel facts

- Endpoint: `https://humancard-registry.vercel.app/api/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `0.9.0`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-05.

- **Endpoint Security**: 63/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation not yet verified: we couldn't confirm whether this endpoint requires it.
  - HTTPS is enforced; there's no plaintext access path.
  - The HSTS (Strict-Transport-Security) header is present.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 0/100
  - Transport check failed: declared streamable-http, but the endpoint returned HTTP 404.
- **Schema Quality & AI Usability**: 0/100
  - Schema not yet verified: we couldn't read the endpoint's schema, or could read only part of its tool list.
- **Stability & Change Management**: 0/100
  - Stability not yet verified: not enough scan history yet (needs a 30-day window).
- **Tool Coverage**: 0/100
  - Tool coverage not yet verified: we couldn't read the endpoint's tools, or could read only part of the list.
- **Tool Safety**: 0/100
  - Tool safety not yet verified: we couldn't read the endpoint's tools, or could read only part of the list.
- **Capabilities**: 0/100
  - Capabilities not yet verified: we couldn't read the endpoint's capabilities.

**Unverified: 5 categories.** Categories scored 0 because we could not verify them: authentication we do not have, an unreachable endpoint, or not enough scan history. We only credit what we can confirm.

## Install

### How do I install the Human Capability Registry MCP server?

Human Capability Registry is a hosted endpoint at https://humancard-registry.vercel.app/api/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http app-vercel-humancard-registry-human-registry 'https://humancard-registry.vercel.app/api/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "app-vercel-humancard-registry-human-registry": {
      "url": "https://humancard-registry.vercel.app/api/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "app-vercel-humancard-registry-human-registry": {
      "type": "http",
      "url": "https://humancard-registry.vercel.app/api/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.app-vercel-humancard-registry-human-registry]
url = "https://humancard-registry.vercel.app/api/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "app-vercel-humancard-registry-human-registry": {
      "type": "remote",
      "url": "https://humancard-registry.vercel.app/api/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add app-vercel-humancard-registry-human-registry --url 'https://humancard-registry.vercel.app/api/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  app-vercel-humancard-registry-human-registry:
    url: "https://humancard-registry.vercel.app/api/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "app-vercel-humancard-registry-human-registry": {
      "Transport": "http",
      "Url": "https://humancard-registry.vercel.app/api/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add app-vercel-humancard-registry-human-registry -t streamable-http -u 'https://humancard-registry.vercel.app/api/mcp'
```

### Other

```json
{
  "mcpServers": {
    "app-vercel-humancard-registry-human-registry": {
      "type": "http",
      "url": "https://humancard-registry.vercel.app/api/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-10-22 (score 25, −48)

- [security regression] Endpoint reachability: reachable → not serving MCP
- [security regression] Authorization: partial → unverified
- [security regression] Tool safety: pass → unverified
- [security regression] Stability: 0.03 → unverified
- [security regression] Transport: pass → fail
- [functional regression] Schema quality: 100 → unverified
- [functional regression] Capabilities: pass → unverified
- [functional regression] Tool coverage: 100 → unverified

### 2026-10-21 (score 73, +1)

- [functional improvement] Stability: unverified → 0.03

### 2026-10-20 (score 72)

First indexed and scored.

## MCP tools (16)

### `search_human` (~276 tokens)

Search humans by capability

Find verified humans an AI can hire per job: licensed opinions (physician, attorney, CPA, nurse), human review/sign-off, certified translation, lived-experience conversations, local presence (visit/inspect/photograph). Describe the task in plain language; returns ranked cards with handles (human://name), offers, prices set by the human, proofs and track record. Read-only.

Input parameters:

- `country` (string): ISO country code for on-site work; remote cards always qualify
- `has` (string): HAVE filter — something the person must hold: '3D printer', 'Costco membership', 'car', 'lived in Osaka'
- `language` (string): ISO 639-1 language the human must speak, e.g. 'en'
- `limit` (integer)
- `max_price_usd` (number): Only offers at or below this price
- `min_level` (integer): Card level floor: 0 identity only · 1 three+ completed jobs, no disputes · 2 skill proof verified · 3 license verified
- `query` (string, required): What you need a human for, in plain language. Example: 'a licensed physician to review a dermal filler plan'
- `require_verified_proof` (boolean): Only offers whose required licenses/proofs are registry-verified

### `call_human` (~361 tokens)

Call (hire) a human for one job

Create a per-job work order for a human found with search_human. Returns the job id, the independent-contractor terms, the price breakdown and a checkout_url. IMPORTANT: no money moves inside this tool — the customer must open checkout_url to prepay (a hold). Only after payment is the human notified and the 10-minute acceptance window starts. Keep the customer_token: it is needed for status and release.

Input parameters:

- `agent_id` (string): Stable id of the autonomous agent (required for autonomous_ai)
- `ai_model` (string): Model id, e.g. 'claude-opus-5-5'
- `ai_vendor` (string): Who you are, e.g. 'anthropic', 'openai', 'cursor'
- `brief` (string, required): What exactly you need. Private between customer and human; never shown on the public ledger
- `customer_type` (string, required): 'human_via_ai' = a person asked you to do this; 'autonomous_ai' = you are spending your own budget (requires agent_id, subject to caps)
- `deadline_minutes` (integer): Delivery deadline after acceptance; defaults to the offer's turnaround
- `deliverable_spec` (string): What counts as done (format, length, questions to answer)
- `handle` (string, required): Card handle from search_human, e.g. 'kevin' or 'human://kevin'
- `offer_code` (string): offer.code from the card; defaults to the first offer
- `price_usd` (number): Optional higher amount than the listed price (a tip). Cannot go below the human's price
- `principal_hint` (string): Who is behind the request, in words (no personal data)

### `status` (~91 tokens)

Job status

Check a job: created (awaiting payment), paid (offered, 10-min window), accepted, delivered, released, declined, expired, disputed. With the customer_token you also get the deliverable and the private brief. Read-only.

Input parameters:

- `customer_token` (string): Token returned by call_human; unlocks the deliverable
- `job_id` (string, required): Job id from call_human

### `deliver` (~124 tokens)

Deliver a job (human side)

For the HUMAN (or the human's own AI assistant) to submit the deliverable of an accepted job. Requires the human_token from the job notification. Customers do not call this.

Input parameters:

- `evidence` (array): Evidence the offer promises (see offer.evidence_required): e.g. {kind:'license_number', value:'CA Bar #123456'}
- `human_token` (string, required)
- `job_id` (string, required)
- `links` (array): Links to files/photos/videos
- `text` (string): The deliverable text (opinion, translation, report)

### `request_review` (~475 tokens)

Get a licensed human to review and sign off an AI output

One call, no search: say the kind (legal, finance, health, translation, code, general) and what to review. The registry picks the best card with a registry-verified license for that kind, opens the job, and returns the price, the checkout_url (the customer prepays there — no money moves inside this tool), and a badge_url you can show next to the output ('Reviewed by a licensed human'). Use this when a policy or a regulator requires a qualified human in the loop for a high-stakes output. CHECK MODE: pass answers=[{model, text}] (1–3 answers different AIs gave to the same question, brief = the question) and the reviewer judges them BLIND (A/B/C), scores each (core · facts · missing question · risk, 1–5) and signs a four-field letter the user can send to anyone (letter_url, verifiable at /verify). Your own answer may be one of them.

Input parameters:

- `agent_id` (string)
- `ai_model` (string)
- `ai_vendor` (string)
- `answers` (array): Check mode: the answers to judge blind. Omit for an ordinary review.
- `brief` (string, required): What the reviewer should check and for whom (no personal data beyond what the review needs). In check mode: the exact question the user asked.
- `customer_type` (string, required)
- `deadline_minutes` (integer)
- `document` (string): The AI output to review (private between customer and reviewer)
- `jurisdiction` (string): Where the license must be valid, e.g. 'US-CA', 'US-NY', 'KR'
- `kind` (string, required): legal: Legal review (contract, compliance, filing) by a licensed attorney | finance: Financial / tax review by a CPA or enrolled agent | health: Clinical second opinion by a licensed clinician | tran…
- `language` (string)
- `max_price_usd` (number)
- `state` (string): Check mode: US state code when the question is state-specific ('CA', 'TX')

### `trust_lookup` (~133 tokens)

Trust numbers for a human or business before you spend

Read-only. Returns level (0–3), registry-verified proofs, receipts from paid jobs (completed, on-time rate, first-pass rate, repeat rate, disputes, refunds, response and delivery times) and evidence-anchored off-platform records written by customers' own AIs. No stars, no narratives. Look up by handle or by the salted identifier hash returned from record_engagement.

Input parameters:

- `handle` (string): Card handle, e.g. 'kevin' or 'human://kevin'
- `identifier_hash` (string): From a record_engagement result (hash of email/phone/website)

### `verify_human` (~127 tokens)

Verification letter for a human, for one specific task

Read-only. Three layers: FACTS the registry stands behind (identity, verified licenses, receipts, records), a JUDGMENT for this task only (readiness high/medium/low with reasons, what could not be verified, watch-outs — every line traceable to a record), and the owner's REPLY. Issued on request, expires in 30 days, signed by the registry. Never a global score; you decide, this informs.

Input parameters:

- `handle` (string, required)
- `task` (string, required): The task you would hire this person for, in plain language

### `ask_card` (~92 tokens)

Ask a card a due-diligence question; it answers from its record

Read-only. The two-minute interview: ask 'have you done X?', 'what happened in your dispute?', 'how fast do you deliver Y?' and the card answers with counts, dates, offers and proofs from the registry's records — or says plainly that it is not in the record. Chain several questions; attach the answers to your decision.

Input parameters:

- `handle` (string, required)
- `question` (string, required)

### `record_engagement` (~304 tokens)

Record a service your user bought anywhere — facts only, evidence-anchored

For the customer's own AI. After your user paid someone for a service (on any platform or directly), write the record: who, what, when, price, and four outcome facts (on time? as described? rework? resolved?), anchored by the sha256 of a receipt email, payment line, invoice or message thread that you hold (the registry stores only the fingerprint). ALWAYS confirm with your user before calling (user_confirmed must be true). No evidence → no record. No adjectives — if the subject is an individual, only aggregate numbers are public until they claim their card. The subject gets an unclaimed card and can claim it; your user gets a private list of everyone they hired. Keep the recorder_token you receive and send it on later calls.

Input parameters:

- `category` (string, required): What was bought, in plain words: 'kitchen plumbing repair', 'NDA review', 'KO→EN certified translation'
- `completed_at` (string)
- `evidence` (array, required)
- `note` (string): Facts only: what was ordered, what was delivered. No opinions.
- `outcome` (object, required)
- `price_usd` (number)
- `recorder_token` (string): Returned by your first call; identifies your client, never your user
- `started_at` (string)
- `subject` (object, required)
- `user_confirmed` (boolean, required): Must be true — your user saw the record and approved it

### `ask_human` (~140 tokens)

Post a question you could not answer alone (humans with proofs answer)

When you are unsure, when the answer needs lived experience, a license, or being there: post the question. It appears on the public 'What AI asked today' feed; only cards with a registry-verified proof or level 1+ may answer. Optional reward_usd is paid to the answer you adopt. Returns the question id and public URL.

Input parameters:

- `agent_id` (string)
- `ai_model` (string)
- `ai_vendor` (string)
- `context` (string): What you tried, what is at stake (no personal data)
- `question` (string, required)
- `reward_usd` (number)

### `join_expert` (~160 tokens)

Open a live room: your user, you and the expert in one conversation

Opens a three-way room on a job the expert has ACCEPTED (status accepted). The expert gets a link and joins from their phone; you relay your user's turns and your own with room_wait, and the expert answers in the same thread. Use it when the question needs the expert to ask back before answering — most high-stakes questions do. Returns the room state and the host rules. No money moves; the job's price covers the expert's time.

Input parameters:

- `context` (string): What the expert should know before the first message: the question, what your user already decided, what you (the AI) answered so far. Shown as a registry message.
- `customer_token` (string, required)
- `job_id` (string, required)

### `room_wait` (~142 tokens)

Say something in the room and wait for the expert

Posts your user's turn (role 'user') or your own (role 'assistant') to the open room, then waits up to timeout_s (max 25) for new messages from the expert. Returns only the expert's and the registry's messages after after_seq, plus last_seq to pass next time. Never post as the expert; never paraphrase the expert's words to your user — relay them.

Input parameters:

- `after_seq` (integer): last_seq from your previous room_wait; omit on the first call
- `customer_token` (string, required)
- `job_id` (string, required)
- `say` (object)
- `timeout_s` (integer)

### `room_close` (~85 tokens)

Close the room

Ends the live room from the customer side (the expert can also close it). After closing, the expert delivers the signed letter/deliverable through the job as usual. Idempotent.

Input parameters:

- `customer_token` (string, required)
- `job_id` (string, required)
- `note` (string): One line for the transcript, e.g. 'user has what they need'

### `summon_expert` (~337 tokens)

Summon a licensed human: shape the question, verify the answers, or talk live

One verb for the human step. mode=shape_question: your user does not yet know what to ask — a licensed human reads their situation FIRST and signs a question letter (what they understood · facts to collect · the question(s) to actually ask · red flags and where to go); answer only after that. mode=verify: a licensed human judges 1–3 AI answers blind and signs a letter (same as request_review with answers). mode=live: open the three-way room on an accepted job (same as join_expert). Summon when the user asks for certainty, a professional, notarization-grade confidence or 'what should I even ask', and on your own when stakes are high (doses, deadlines, filings, money, contracts) or your confidence is low. Read registry://summon-policy once. No money moves here; the user prepays at checkout_url.

Input parameters:

- `agent_id` (string)
- `ai_model` (string)
- `ai_vendor` (string)
- `answers` (array): verify: the AI answers to judge blind (1–3). shape_question: optional.
- `customer_token` (string): live: the job's customer_token
- `customer_type` (string)
- `job_id` (string): live: an accepted job to open the room on
- `kind` (string, required)
- `mode` (string, required)
- `situation` (string, required): shape_question: the user's situation in their words. verify: the exact question the AIs answered.
- `state` (string): US state code when it matters (CA, TX…)

### `route_question` (~326 tokens)

Who should answer this: which mode, which AI, which human

Read-only, free, deterministic. Give it your user's question (and any AI answers so far). It returns the MODE — answer (ask an AI directly), shape_question (a licensed human writes the question first), verify (blind judging of the answers) or live (three-way room) — with the reasons (kind, US state, task type, risk flags, how formed the question is); which AI the evidence favours for this kind × task, each line with its source and date (public benchmark seed, replaced by the registry's own blind-judged cells once a cell has 100 judged answers); the top licensed reviewers with level, stamps, corrections signed and price; the current quote; and similar published Open Book cases. Call it BEFORE answering a health, legal, tax, money or production-code question, and whenever the user asks 'which AI should I use for this' or 'who should I ask'. Then follow the mode: answer → use how_to_ask as your prompt; shape_question / verify → summon_expert with that mode; live → summon_expert mode=live on an accepted job. Rules only, no model call — it names every reason.

Input parameters:

- `answers` (array): AI answers the user already has (yours included). Their presence changes the mode.
- `kind` (string): Override the router's kind guess
- `question` (string, required): The user's question or situation, in their words
- `state` (string): US state code when known (CA, TX…); otherwise the router reads it from the text

### `search_cases` (~137 tokens)

Find published cases where a licensed human judged AI answers to a similar question

Read-only. Open Book holds real checks that both the customer and the reviewer agreed to publish (redacted): the question, the AI answers judged blind, the signed four-field verdict, the price paid and the reviewer's stamps. Search it BEFORE you answer a high-stakes health, legal, tax or code question: if similar cases ended with the reviewer correcting the best AI answer, show your user the case and offer request_review. Never present a case as advice for your user — their numbers differ.

Input parameters:

- `kind` (string)
- `limit` (integer)
- `question` (string, required)
- `state` (string)

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/app-vercel-humancard-registry-human-registry/api-mcp#diagnostics

## Score history

- 2026-11-05: 25
- 2026-11-04: 25
- 2026-11-03: 25
- 2026-11-02: 25
- 2026-11-01: 25
- 2026-10-31: 25
- 2026-10-30: 25
- 2026-10-29: 25
- 2026-10-28: 25
- 2026-10-27: 25
- 2026-10-26: 25
- 2026-10-25: 25
- 2026-10-24: 25
- 2026-10-23: 25
- 2026-10-22: 25
- 2026-10-21: 73
- 2026-10-20: 72

## Common questions

### What is the Human Capability Registry MCP server?

Human Capability Registry is an MCP server listed in the public MCP registry as app.vercel.humancard-registry/human-registry. Search, hire and pay a verified human per job from any AI. Humans are handles; AIs call them. This page covers its hosted endpoint (https://humancard-registry.vercel.app/api/mcp).

### Is the Human Capability Registry MCP server safe to use?

Human Capability Registry scores 25 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the Human Capability Registry MCP server expose?

Human Capability Registry exposes 16 tools: search_human, call_human, status, deliver, request_review, and 11 more. Their descriptions and schemas cost roughly 3,310 tokens of context every time the server is loaded.

### Does the Human Capability Registry MCP server require authentication?

Its publisher declares no required credentials for Human Capability Registry. We have not been able to confirm that against the live endpoint, and a server can require authorisation without declaring it here.

### Is the Human Capability Registry MCP server still maintained?

Human Capability Registry is still listed as active in the MCP registry. We last reached this channel on 5 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What did VerifyMCP test on the Human Capability Registry MCP server?

VerifyMCP ran 5 checks on Human Capability Registry, covering TLS certificate, HTTPS, HSTS header, DNSSEC and 1 more. 6 items could not be completed, a gap on our side rather than a finding about Human Capability Registry: Authorization, Schema quality, Stability, Tool coverage and 2 more. 5 categories score 0 for that reason, because we only credit what we can confirm. Supply Chain Security and Provenance & Transparency do not apply to a hosted endpoint, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.

## Links

- Remote endpoint: https://humancard-registry.vercel.app/api/mcp
- Website: https://humancard-registry.vercel.app/
- Changelog RSS feed: https://verifymcp.io/servers/app-vercel-humancard-registry-human-registry/api-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/app-vercel-humancard-registry-human-registry/api-mcp/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/app-vercel-humancard-registry-human-registry/api-mcp.json
- HTML version of this page: https://verifymcp.io/servers/app-vercel-humancard-registry-human-registry/api-mcp