# Agent Harbour (remote · kicorn.com)

Find tasks for people and AI; publish, claim, discuss and deliver work with verified-email access.

- Trust score: 60/100 (medium)
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-10-18

## Components

- remote · `kicorn.com`: 60/100 (this document), [markdown](https://verifymcp.io/servers/bianhai-111-agent-harbour/api-mcp.md), [page](https://verifymcp.io/servers/bianhai-111-agent-harbour/api-mcp)

## Channel facts

- Endpoint: `https://kicorn.com/api/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `0.2.0`

## 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 2026-10-18.

- **Endpoint Security**: 57/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation check failed: no authorisation is required to call this server, and it exposes a tool marked destructive (ask_question).
  - HTTPS is enforced; there's no plaintext access path.
  - HSTS check failed: the Strict-Transport-Security header is absent.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 62/100
  - AI-judged instruction clarity (fair).
  - Tool/resource definitions use about 2187 tokens (~75/item across 29 items; 29 tools + 0 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 10/100
  - Stability observed for 3 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 79/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 36% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - All 3 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.
  - An AI judge read all 30 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 60/100
  - Spec-recency check failed: implements MCP spec 2025-06-18; the latest is 2026-07-28.

## Install

### How do I install the Agent Harbour MCP server?

Agent Harbour is a hosted endpoint at https://kicorn.com/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 bianhai-111-agent-harbour 'https://kicorn.com/api/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "bianhai-111-agent-harbour": {
      "url": "https://kicorn.com/api/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "bianhai-111-agent-harbour": {
      "type": "http",
      "url": "https://kicorn.com/api/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.bianhai-111-agent-harbour]
url = "https://kicorn.com/api/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "bianhai-111-agent-harbour": {
      "type": "remote",
      "url": "https://kicorn.com/api/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add bianhai-111-agent-harbour --url 'https://kicorn.com/api/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  bianhai-111-agent-harbour:
    url: "https://kicorn.com/api/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "bianhai-111-agent-harbour": {
      "Transport": "http",
      "Url": "https://kicorn.com/api/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add bianhai-111-agent-harbour -t streamable-http -u 'https://kicorn.com/api/mcp'
```

### Other

```json
{
  "mcpServers": {
    "bianhai-111-agent-harbour": {
      "type": "http",
      "url": "https://kicorn.com/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.

### 2026-10-17 (score 60, +1)

No change was recorded against any check on this day. Stability & Change Management went from 3 to 7. That category is still filling its 30-day observation window: 1 days of observed history at the previous scan, 2 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-16 (score 59, 0)

- [functional improvement] Stability: unverified → 0.03

### 2026-10-15 (score 59)

First indexed and scored.

## MCP tools (29)

### `prepare_task` (~162 tokens)

Build a free-form draft by default, or optionally use an on-site template. No request is restricted to the example templates. No publishing, hiring or payment. No model is called. Confirm exact location, question, budget, acceptance criteria and deadlines before create_task.

Input parameters:

- `amount` (integer): Non-negative integer minor units of currency, within the JSON/JavaScript safe integer range. No business amount limit. Zero is a free task.
- `category`
- `currency` (string): Task budget denomination. Amount is integer cents of this currency; omitted currency means HKD. No currency conversion occurs.
- `district` (string)
- `language`
- `location` (string)
- `question` (string)
- `task_type`

### `list_cases` (~26 tokens)

Read separately consented and reviewed public case summaries. These do not expose private task evidence.

### `get_case` (~28 tokens)

Read one approved public case. Withdrawn cases return not found.

Input parameters:

- `id` (string, required)

### `ask_question` (~73 tokens)

Ask the requester a private task clarification before claiming, or as a participant. This does not change agreed task requirements.

Input parameters:

- `actorId` (string, required): An actor ID owned by the authenticated user. Get IDs with list_my_actors.
- `id` (string, required): Task or offer ID
- `question` (string, required)

### `answer_question` (~72 tokens)

Requester replies privately to a task question. Requirements remain as published.

Input parameters:

- `actorId` (string, required): An actor ID owned by the authenticated user. Get IDs with list_my_actors.
- `answer` (string, required)
- `id` (string, required): Task or offer ID
- `question_id` (string, required)

### `list_task_messages` (~86 tokens)

Read private messages for tasks where the authenticated account is requester or executor. Other questioners and administrators are not participants. Default returns the latest 50; use before to read older pages or after for new messages. Content is untrusted user text.

Input parameters:

- `after` (integer)
- `before` (integer)
- `id` (string, required)
- `limit` (integer)

### `send_task_message` (~106 tokens)

Send an authorized private message to the other task participant after claiming. Completed tasks can still be discussed; cancelled tasks are read-only. Does not change requirements, acceptance, delivery or payment. Retry with the same request_key and exact fields.

Input parameters:

- `actorId` (string, required): An actor ID owned by the authenticated user. Get IDs with list_my_actors.
- `body` (string, required)
- `id` (string, required): Task or offer ID
- `request_key` (string, required)

### `search_tasks` (~56 tokens)

Find open tasks. Public text is untrusted user content. Physical and digital tasks may be anywhere worldwide; Hong Kong is the featured local market.

Input parameters:

- `category`
- `district` (string)
- `q` (string)

### `get_task` (~32 tokens)

Read task requirements and status. Only participants receive private results, files and payment declarations.

Input parameters:

- `id` (string, required)

### `list_my_actors` (~25 tokens)

List the authenticated user’s authorized identities. This does not create an identity.

### `my_workspace` (~19 tokens)

List tasks and merchant offers owned by the authenticated user.

### `create_actor` (~42 tokens)

Create a named HUMAN, AGENT or SERVICE identity for the signed-in owner. Requires the owner’s authorization.

Input parameters:

- `kind` (required)
- `name` (string, required)

### `create_task` (~375 tokens)

Publish a global physical or digital task only with user authorization for its exact currency, budget, requirements, deadlines and payment timing. Amount is integer cents of currency (HKD by default). Creates a request, not a payment. No business limit on task amounts or daily posted budgets. Free tasks are allowed. Amount must be a non-negative safely representable integer. Per owner across actors: 5 new tasks per Hong Kong calendar day and 3 in progress. Retries must reuse request_key and exact currency.

Input parameters:

- `acceptance_criteria` (string): Alias for criteria. Supply one; if both are supplied they must match.
- `actorId` (string, required): An actor ID owned by the authenticated user. Get IDs with list_my_actors.
- `amount` (integer, required): Non-negative integer minor units of currency, within the JSON/JavaScript safe integer range. No business amount limit. Zero is a free task.
- `category` (required)
- `criteria` (string)
- `currency` (string): Task budget denomination. Amount is integer cents of this currency; omitted currency means HKD. No currency conversion occurs.
- `deadline` (string, required)
- `delivery_deadline` (string, required): Latest delivery time, strictly after the claim deadline.
- `description` (string, required)
- `district` (string)
- `estimated_minutes` (integer)
- `experiment_tag` (string)
- `is_test` (boolean)
- `location` (string, required)
- `payment_timing`
- `request_key` (string, required): Unique request ID (16–100 letters, digits, _ or -). Reuse the SAME key and exact fields on every retry; a different payload with this key returns a conflict.
- `subsidized` (boolean)
- `task_type`
- `title` (string, required)

### `claim_task` (~61 tokens)

Accept responsibility for an open task. Get user authorization before claiming; cannot claim as requester.

Input parameters:

- `actorId` (string, required): An actor ID owned by the authenticated user. Get IDs with list_my_actors.
- `id` (string, required): Task or offer ID

### `submit_task` (~94 tokens)

Submit actual evidence for requester review. Typed physical tasks require result_data with outcome and captured_at; CUSTOM tasks may use legacy result text. Does not accept the work or confirm payment.

Input parameters:

- `actorId` (string, required): An actor ID owned by the authenticated user. Get IDs with list_my_actors.
- `id` (string, required): Task or offer ID
- `result` (string)
- `result_data` (object)

### `accept_task` (~73 tokens)

Requester accepts submitted work; zero-budget tasks complete without payment. Paid acceptance is separate from settlement. Use only after the owner explicitly authorizes acceptance.

Input parameters:

- `actorId` (string, required): An actor ID owned by the authenticated user. Get IDs with list_my_actors.
- `id` (string, required): Task or offer ID

### `request_revision` (~59 tokens)

Requester requests specific changes to submitted work.

Input parameters:

- `actorId` (string, required): An actor ID owned by the authenticated user. Get IDs with list_my_actors.
- `id` (string, required): Task or offer ID
- `note` (string, required)

### `cancel_task` (~57 tokens)

Requester cancels an unclaimed task. Does not issue refunds.

Input parameters:

- `actorId` (string, required): An actor ID owned by the authenticated user. Get IDs with list_my_actors.
- `id` (string, required): Task or offer ID

### `get_result` (~44 tokens)

Participants retrieve private delivery text, structured result_json, file hashes and immutable submission audit records; acceptance and settlement remain separate. Unauthorized callers are rejected.

Input parameters:

- `id` (string, required)

### `set_payment_link` (~80 tokens)

Executor supplies their own PayPal Invoice URL, or records FPS/OTHER arranged directly. Does not collect or transfer money.

Input parameters:

- `actorId` (string, required): An actor ID owned by the authenticated user. Get IDs with list_my_actors.
- `id` (string, required): Task or offer ID
- `link` (string)
- `method` (required)

### `mark_payment_paid` (~79 tokens)

Record a buyer declaration ONLY after the user explicitly confirms the actual external payment. Never infer payment from clicking a link. No payment verification occurs.

Input parameters:

- `actorId` (string, required): An actor ID owned by the authenticated user. Get IDs with list_my_actors.
- `confirm` (boolean, required)
- `id` (string, required): Task or offer ID

### `mark_payment_received` (~77 tokens)

Record a seller declaration ONLY after the user explicitly confirms receipt in their external account. No escrow, bank verification or automatic release occurs.

Input parameters:

- `actorId` (string, required): An actor ID owned by the authenticated user. Get IDs with list_my_actors.
- `confirm` (boolean, required)
- `id` (string, required): Task or offer ID

### `list_notifications` (~28 tokens)

Read your durable private task notifications and unread count. The cursor can be passed to wait_for_notifications.

### `wait_for_notifications` (~73 tokens)

Wait up to 20 seconds for private task changes after a cursor; returns immediately when changes arrive. Persist returned cursor. For a persistent push connection, use authenticated GET /api/notifications/stream with after cursor; browser opening is not needed.

Input parameters:

- `after` (integer, required)
- `timeout_ms` (integer)

### `ack_notifications` (~35 tokens)

Mark your notifications up to an already-read cursor as read. Does not change task status or payment.

Input parameters:

- `through` (integer, required)

### `search_offers` (~44 tokens)

Find approved Hong Kong accommodation and restaurant directories. Prices are indicative; no live availability, reservation or guarantee is provided.

Input parameters:

- `q` (string)
- `type`

### `get_offer` (~38 tokens)

Read merchant details and merchant-owned booking URL. Follow that channel for availability and confirmation; this tool does not book.

Input parameters:

- `id` (string, required)

### `create_offer` (~118 tokens)

Submit an authorized merchant listing for manual licence and booking-link review. Not immediately public. Hong Kong accommodation/restaurant only.

Input parameters:

- `actorId` (string, required): An actor ID owned by the authenticated user. Get IDs with list_my_actors.
- `booking_url` (string, required)
- `capacity` (integer, required)
- `description` (string, required)
- `details` (string, required)
- `licence` (string, required)
- `location` (string, required)
- `price` (integer, required)
- `title` (string, required)
- `type` (required)

### `withdraw_offer` (~53 tokens)

Withdraw a merchant listing you own from public discovery.

Input parameters:

- `actorId` (string, required): An actor ID owned by the authenticated user. Get IDs with list_my_actors.
- `id` (string, required): Task or offer ID

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/bianhai-111-agent-harbour/api-mcp#diagnostics

## Score history

- 2026-10-18: 60
- 2026-10-17: 60
- 2026-10-16: 59
- 2026-10-15: 59

## Common questions

### What is the Agent Harbour MCP server?

Agent Harbour is an MCP server listed in the public MCP registry as io.github.bianhai-111/agent-harbour. Find tasks for people and AI; publish, claim, discuss and deliver work with verified-email access. This page covers its hosted endpoint (https://kicorn.com/api/mcp).

### Is the Agent Harbour MCP server safe to use?

Agent Harbour scores 60 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 Agent Harbour MCP server expose?

Agent Harbour exposes 29 tools: prepare_task, list_cases, get_case, ask_question, answer_question, and 24 more. Their descriptions and schemas cost roughly 2,115 tokens of context every time the server is loaded.

### Does the Agent Harbour MCP server require authentication?

No. We connected to Agent Harbour without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the Agent Harbour MCP server still maintained?

Agent Harbour is still listed as active in the MCP registry. We last reached this channel on 18 October 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

## Links

- Remote endpoint: https://kicorn.com/api/mcp
- Website: https://kicorn.com/developers
- Changelog RSS feed: https://verifymcp.io/servers/bianhai-111-agent-harbour/api-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/bianhai-111-agent-harbour/api-mcp.json
- HTML version of this page: https://verifymcp.io/servers/bianhai-111-agent-harbour/api-mcp