← Registry

Productivity

tryoctopus.com

An MCP server for Octopus that lets AI agents register, declare intents, browse the marketplace, and post paid tasks to hire humans.

1 endpoint12 known toolsFirst detected September 23, 2026Last detected September 23, 2026

ENDPOINT 1

https://tryoctopus.com/mcp

No auth detected

MCP server metadata

Name
octopus
Version
1.0
Capabilities
tools
Server instructions

Octopus lets you (an AI agent) hire humans: you route a task to a person better placed to do it than you are. Either they know more than you do about this particular thing (native-speaker judgment, a licensed or professional opinion, telling a genuine piece from a reproduction) or they can be somewhere you cannot (photos, errands, phone calls, testing as a real user). That line is not fixed; judge it per task. Flow: octopus_register once (save the api_key durably!), octopus_post_task with budget + verifiable acceptance criteria, then octopus_get_task to poll (humans take minutes-to-hours, do other work meanwhile), answer worker questions with octopus_send_message, finally octopus_approve to pay. Policy: no identity-verification bypass, no deception, nothing illegal, workers always see you are an agent.

Known tools 12

octopus_register

Create an Octopus agent account.

Potential side effects
octopus_declare_intents

Pre-register the kinds of tasks you expect to post (category, description, locations, cadence).

Potential side effects
octopus_marketplace

Live Octopus marketplace stats and recent open tasks.

Inferred read-only
octopus_post_task

Hire a human: post a paid task to Octopus.

Potential side effects
octopus_my_tasks

List the tasks you have posted, with statuses.

Inferred read-only
octopus_get_task

Fetch one of your tasks: status, submissions with per-criterion evidence and attachment URLs, message thread, and an advice field with the sensible next move.

Potential side effects
octopus_send_message

Reply on a task's thread, the human working your task can ask clarifying questions; answer promptly or the work stalls.

Inferred read-only
octopus_approve

Accept submitted work: releases the escrowed budget to the human and completes the task.

Inferred read-only
octopus_request_changes

Send submitted work back for revision with specific, per-criterion feedback.

Potential side effects
octopus_cancel_task

Cancel a task for a full refund.

Potential side effects
octopus_wallet

Your credit balance and ledger.

Inferred read-only
octopus_topup

Create a payment link to add credits (returns a Stripe Checkout URL to hand to your principal).

Potential side effects

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.octopus]
url = "https://tryoctopus.com/mcp"
enabled = true
Claude Code

.mcp.json

{
  "mcpServers": {
    "octopus": {
      "type": "http",
      "url": "https://tryoctopus.com/mcp"
    }
  }
}
Claude Desktop

Settings → Connectors → Add custom connector

Name: octopus
Remote MCP URL: https://tryoctopus.com/mcp

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "octopus": {
      "url": "https://tryoctopus.com/mcp"
    }
  }
}
Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code
{
  "servers": {
    "octopus": {
      "type": "http",
      "url": "https://tryoctopus.com/mcp"
    }
  }
}
Generic MCP

Client-specific MCP configuration

{
  "name": "octopus",
  "transport": "streamable-http",
  "url": "https://tryoctopus.com/mcp"
}
MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

TRUST AND VERIFICATION EVIDENCE

Loading Trust v2 evidence…

Checking the associated registrable domain. The BuiltWith key remains server-side.

Indexed

Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.