# askingahuman (npm · @askingahuman/mcp-server)

Let AI agents reach a person they choose: free questions and paid live text chats.

- Trust score: 73/100 (medium)
- Change this week: +3
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- npm · `@askingahuman/mcp-server`: 73/100 (this document), [markdown](https://verifymcp.io/servers/com-askingahuman-mcp-server/askingahuman-mcp-server.md), [page](https://verifymcp.io/servers/com-askingahuman-mcp-server/askingahuman-mcp-server)

## Channel facts

- Registry: `npm`
- Package: `@askingahuman/mcp-server`
- Version: `2.2.0`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, 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.

- **Supply Chain Security**: 98/100
  - No malware found by supply-chain analysis.
  - No known CVEs affecting this package version or its production dependencies.
  - No install/post-install scripts declared.
  - 40 of 130 dependencies flagged as unhealthy.
- **Provenance & Transparency**: 19/100
  - Repository check failed: no source repository is declared.
  - Provenance check failed: no build-provenance attestation is published.
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 17 days ago).
  - Security-disclosure policy not yet verified: we couldn't inspect the source repository.
- **Schema Quality & AI Usability**: 76/100
  - AI-judged instruction clarity (excellent).
  - Context-footprint check failed: tool/resource definitions use about 1765 tokens (~126/item across 14 items; 14 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 70/100
  - Stability observed for 21 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
- **Tool Safety**: 75/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - 0 of 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "send_message" implies "send" and declares no destructiveHint at all, which the MCP spec reads as destructive by default.
  - An AI judge read all 14 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### How do I install the askingahuman MCP server?

askingahuman runs locally as an npm package, launched with npx -y @askingahuman/mcp-server. 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 com-askingahuman-mcp-server -- npx -y @askingahuman/mcp-server
```

### Cursor

```json
{
  "mcpServers": {
    "com-askingahuman-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "@askingahuman/mcp-server"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "com-askingahuman-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "@askingahuman/mcp-server"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add com-askingahuman-mcp-server -- npx -y @askingahuman/mcp-server
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "com-askingahuman-mcp-server": {
      "type": "local",
      "command": [
        "npx",
        "-y",
        "@askingahuman/mcp-server"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add com-askingahuman-mcp-server --command npx --arg -y --arg @askingahuman/mcp-server
```

### Hermes

```yaml
mcp_servers:
  com-askingahuman-mcp-server:
    command: "npx"
    args: ["-y", "@askingahuman/mcp-server"]
```

### Netclaw

```json
{
  "McpServers": {
    "com-askingahuman-mcp-server": {
      "Transport": "stdio",
      "Command": "npx",
      "Arguments": [
        "-y",
        "@askingahuman/mcp-server"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add com-askingahuman-mcp-server -t stdio -c npx -a -y @askingahuman/mcp-server
```

### Other

```json
{
  "mcpServers": {
    "com-askingahuman-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "@askingahuman/mcp-server"
      ]
    }
  }
}
```

## 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.

### 2026-11-04 (score 73, +1)

No change was recorded against any check on this day. Stability & Change Management went from 63 to 67. That category is still filling its 30-day observation window: 19 days of observed history at the previous scan, 20 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-02 (score 72, +1)

No change was recorded against any check on this day. Stability & Change Management went from 57 to 60. That category is still filling its 30-day observation window: 17 days of observed history at the previous scan, 18 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-31 (score 71, +1)

No change was recorded against any check on this day. Stability & Change Management went from 50 to 53. That category is still filling its 30-day observation window: 15 days of observed history at the previous scan, 16 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-29 (score 70, +1)

No change was recorded against any check on this day. Stability & Change Management went from 43 to 47. That category is still filling its 30-day observation window: 13 days of observed history at the previous scan, 14 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-26 (score 69, +1)

No change was recorded against any check on this day. Stability & Change Management went from 33 to 37. That category is still filling its 30-day observation window: 10 days of observed history at the previous scan, 11 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-24 (score 68, +1)

No change was recorded against any check on this day. Stability & Change Management went from 27 to 30. That category is still filling its 30-day observation window: 8 days of observed history at the previous scan, 9 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-22 (score 67, +1)

No change was recorded against any check on this day. Stability & Change Management went from 20 to 23. That category is still filling its 30-day observation window: 6 days of observed history at the previous scan, 7 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-20 (score 66, +1)

No change was recorded against any check on this day. Stability & Change Management went from 13 to 17. That category is still filling its 30-day observation window: 4 days of observed history at the previous scan, 5 at this one. The score rises as the window fills, whether or not the server changes.

## MCP tools (14)

### `ask_human` (~274 tokens)

Use this tool ONLY when you encounter a subjective decision, require real-time human verification, or lack the permissions to proceed. You MUST provide a concise context of the problem and the specific question you need the user to answer. Asking is FREE. The question goes to the person you choose (human_id from list_humans), who answers when they are free; not everyone takes free questions. Do not include links or credentials. Returns a questionId — use check_answer to poll for the response.

Input parameters:

- `contact_email` (string): Optional email address to receive a notification when the question is answered.
- `human_id` (string, required): The person to ask (humanId from list_humans or who_is_live).
- `question` (string, required): The question to ask the human. Include current context and your specific goal. Be specific and concise (max 400 characters).
- `screenshot_base64` (string): Optional base64-encoded screenshot image (PNG/JPEG/WebP/GIF, max 5MB) to provide visual context to the human answerer. IMPORTANT: This screenshot is for human eyes only. You (the AI) must NOT process…

### `check_answer` (~81 tokens)

Check if a human has answered your question at askingahuman.com. Poll this tool periodically after calling ask_human — recommended interval is every 30 minutes or less. An expert answers when they are free. Returns the answer text if ready, or the current status if still pending.

Input parameters:

- `question_id` (string, required): The questionId returned by ask_human.

### `reply_to_human` (~99 tokens)

Continue the conversation with the human who answered your question at askingahuman.com: ask a follow-up, clarify, or say thanks. Only works after the question is answered. The human is notified on their phone. Use check_answer to read their response.

Input parameters:

- `message` (string, required): Your message to the human (max 2000 characters).
- `question_id` (string, required): The questionId returned by ask_human (the answered question).

### `list_humans` (~163 tokens)

List the reviewed experts on askingahuman.com: name, title, specializations, and track record (hours live, how fast they answer, how often they accept, and how often agents gave a thumbs up). Every request names a person: use a humanId from here with get_quote, connect_human or ask_human. Profiles are public summaries only; no contact details are shared.

Input parameters:

- `credential` (string): Optional: only experts with a credential of this type that askingahuman has checked.
- `limit` (number): How many to return (1-25, default 10).
- `specialty` (string): Optional: only experts with this specialization.
- `verified_only` (boolean): Optional: only experts with at least one checked credential.

### `who_is_live` (~105 tokens)

See which people are live right now and can take a live chat. Free; nobody is notified. Set wait_seconds to wait (up to 50 seconds) for someone to come online. Use get_quote with a humanId to see their price before connect_human.

Input parameters:

- `specialization` (string): Optional: only people with this specialization.
- `wait_seconds` (number): Wait up to this many seconds (0-50) for someone to be live. Default 0.

### `get_quote` (~120 tokens)

Check, at no cost, what a live chat with a specific person would cost: their price per hour of chat, what is held on the card, how long the chat can run with extensions, and whether they are live now. Nothing is charged and nobody is notified. Call this before connect_human.

Input parameters:

- `human_id` (string, required): The person (humanId from list_humans or who_is_live).
- `max_spend_usd` (number): Optional: the most to spend on the whole chat, extensions included, in US dollars.

### `rate_human` (~131 tokens)

After a live chat or video session ends, or after a free question is answered, give the person a thumbs up or thumbs down (once). It feeds their public thumbs-up rate, which helps other agents choose. Pass session_id for a live session or question_id for an answer.

Input parameters:

- `comment` (string): Optional: a short reason (up to 500 characters).
- `question_id` (string): The questionId from ask_human.
- `rating` (string, required): up: they helped; down: they did not.
- `session_id` (string): The Session ID from connect_human.

### `submit_feedback` (~68 tokens)

Submit feedback about the askingahuman.com MCP server or service. Use this to report issues, suggest improvements, or provide general feedback from an AI agent.

Input parameters:

- `contact_email` (string): Optional email address for follow-up.
- `feedback` (string, required): The feedback message. Be specific and detailed.

### `connect_human` (~256 tokens)

Request a PAID live text chat with a specific person (human_id). Chat is sold by the hour at the price the person sets: the first hour is held on the card when you request and charged in full once they accept (nothing if they never do); extend_session adds another hour. If you end early, paid time is not refunded; if the person ends the chat or leaves early, you pay only for the time used, and if they never join, nothing. If the person is offline, the request waits up to 24 hours for them. Call get_quote first; the user may be asked to approve the cost, and spending never exceeds the limit set by whoever runs this server. All sales are final. Then call check_connection to wait for the person to accept; you join the chat automatically.

Input parameters:

- `human_id` (string, required): The person to chat with (humanId from list_humans or who_is_live).
- `max_spend_usd` (number): Optional: the most to spend on the whole chat, extensions included, in US dollars.
- `message` (string, required): What you need help with, shown to the person before they accept (max 500 characters). No links or credentials.

### `check_connection` (~125 tokens)

Check a live chat requested with connect_human, optionally waiting up to 50 seconds for the person to accept. When they accept you join the chat automatically; then use read_messages and send_message. Shows how much paid time is left. When the session has ended, shows the amount charged, and whether paid time was released because the person left early or never joined.

Input parameters:

- `session_id` (string, required): The Session ID from connect_human.
- `wait_seconds` (number): Wait up to this many seconds (0-50) for the person to accept. Default 0.

### `read_messages` (~97 tokens)

Read new messages from the person in a live session, waiting up to 50 seconds for one to arrive. Returns only messages you have not read yet, plus notices such as the person leaving or the chat ending, and how much paid time is left.

Input parameters:

- `session_id` (string, required): The Session ID.
- `wait_seconds` (number): Wait up to this many seconds (0-50) for a new message. Default 30.

### `send_message` (~66 tokens)

Send a chat message to the person in a live session you have joined. Plain text only, up to 1000 characters; links and credentials are blocked.

Input parameters:

- `session_id` (string, required): The Session ID.
- `text` (string, required): Your message (max 1000 characters).

### `extend_session` (~91 tokens)

Add one more paid block (an hour of chat) to a live chat that has started, charged in full right away to the same card or balance. The user may be asked to approve the cost. Limited by the session's time and spend limits. Use it when check_connection or read_messages says paid time is running low and you still need the person.

Input parameters:

- `session_id` (string, required): The Session ID.

### `end_session` (~89 tokens)

End a live session and leave the chat. A request the person has not accepted yet is cancelled and its hold released. Once they accepted, every hour bought is charged in full when you end it, so ending early does not refund paid time (if the person had already left, only the time used was charged). Call this when you are done.

Input parameters:

- `session_id` (string, required): The Session ID.

## Diagnostics

Captured diagnostic sections: Provenance, Dependencies. The full working is on the page: https://verifymcp.io/servers/com-askingahuman-mcp-server/askingahuman-mcp-server#diagnostics

## Score history

- 2026-11-05: 73
- 2026-11-04: 73
- 2026-11-03: 72
- 2026-11-02: 72
- 2026-11-01: 71
- 2026-10-31: 71
- 2026-10-30: 70
- 2026-10-29: 70
- 2026-10-28: 69
- 2026-10-27: 69
- 2026-10-26: 69
- 2026-10-25: 68
- 2026-10-24: 68
- 2026-10-23: 67
- 2026-10-22: 67
- 2026-10-21: 66
- 2026-10-20: 66
- 2026-10-19: 65
- 2026-10-18: 65
- 2026-10-17: 65
- 2026-10-16: 53
- 2026-10-15: 64
- 2026-10-14: 23

## Common questions

### What is the askingahuman MCP server?

askingahuman is an MCP server listed in the public MCP registry as com.askingahuman/mcp-server. Let AI agents reach a person they choose: free questions and paid live text chats. This page covers its npm package (@askingahuman/mcp-server).

### Is the askingahuman MCP server safe to use?

askingahuman scores 73 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 5 November 2026. It declares no install or post-install scripts. 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 askingahuman MCP server expose?

askingahuman exposes 14 tools: ask_human, check_answer, reply_to_human, list_humans, who_is_live, and 9 more. Their descriptions and schemas cost roughly 1,765 tokens of context every time the server is loaded.

### Is the askingahuman MCP server still maintained?

askingahuman 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 licence is the askingahuman MCP server under?

askingahuman declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.

### What did VerifyMCP test on the askingahuman MCP server?

VerifyMCP ran 15 checks on askingahuman, covering Malware scan, Known CVEs, Install scripts, Dependency health and 11 more. One item could not be completed, a gap on our side rather than a finding about askingahuman: Security disclosure. Endpoint Security and Transport & Reachability do not apply to an npm package, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.

## Links

- npm package: https://www.npmjs.com/package/@askingahuman/mcp-server
- Socket report: https://socket.dev/npm/package/@askingahuman/mcp-server
- Website: https://askingahuman.com/developers
- Changelog RSS feed: https://verifymcp.io/servers/com-askingahuman-mcp-server/askingahuman-mcp-server.xml
- Changelog JSON feed: https://verifymcp.io/servers/com-askingahuman-mcp-server/askingahuman-mcp-server/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/com-askingahuman-mcp-server/askingahuman-mcp-server.json
- HTML version of this page: https://verifymcp.io/servers/com-askingahuman-mcp-server/askingahuman-mcp-server