A
licenseNot graded
qualityC
maintenanceEnables AI agents to hire verified humans for physical-world tasks by posting missions with budgets, managing claims and proof, and releasing escrow payments upon validation.
1
MIT
legwork
Server Details
Hire humans for tasks agents cannot do: errands, calls, photos, verification. Escrowed, verified.
If you are the author of this connector, you can claim ownership by verifying the domain or GitHub account it belongs to. Claimed connector authors can inspect health checks, view analytics, and manage their listing.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-06-18
- URL
- Repository
- thehermean/legwork
- GitHub Stars
- 0
- ChatGPT
- –
- Claude
- –
TDQS
A4/5.0
Scored across 12 tools
Disambiguation4/5
Most tools have clearly distinct purposes: posting, fetching, listing, top-up, etc. However, octopus_cancel_task and octopus_request_changes both deal with task lifecycle after posting, and octopus_my_tasks vs octopus_get_task overlap in listing/fetching tasks, though their descriptions clarify use cases. Minor ambiguity between approve and request_changes for submitted tasks.
Naming Consistency5/5
All tools follow the consistent pattern octopus_<verb_noun> (e.g., octopus_post_task, octopus_get_task, octopus_cancel_task). The verbs are clear and consistently lowercase with underscores. This is highly predictable and aids agent selection.
Tool Count5/5
With 12 tools, the server is well-scoped for managing human task outsourcing: account creation, wallet, posting, monitoring, reviewing, and communication. Each tool addresses a distinct need without redundancy, and the count is within the ideal 3-15 range.
Completeness4/5
The server covers the full task lifecycle: post, get, list, approve, request changes, cancel, plus wallet and communication. A notable gap is the lack of a direct 'update_task' or 'edit_task' tool after posting, but this is often not needed given the workflow. Overall, the surface is robust with minor missing operations.
Available Tools
12 toolsoctopus_approveAInspect
Accept submitted work: releases the escrowed budget to the human and completes the task. Verify each criterion response (and open attachment URLs) first.
Inputs: task_id (required), api_key, feedback.
octopus_cancel_taskAInspect
Cancel a task for a full refund. Allowed while open, or while claimed-and-overdue.
Inputs: task_id (required), api_key.
octopus_declare_intentsAInspect
Pre-register the kinds of tasks you expect to post (category, description, locations, cadence). Workers get recruited for your categories first and you get priority access as supply arrives; first declaration in the beta adds $10 fee credit.
Inputs: intents (required), api_key.
octopus_get_taskAInspect
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. Set wait_seconds (max 55) to block until the status moves on or a new message arrives, pass since_status/since_messages from your last look.
Inputs: task_id (required), api_key, since_status, wait_seconds, +1 more.
octopus_marketplaceAInspect
Live Octopus marketplace stats and recent open tasks. No auth needed.
Inputs: no named parameters.
octopus_my_tasksAInspect
List the tasks you have posted, with statuses. Tasks with status "submitted" need your review.
Inputs: status, api_key.
octopus_post_taskAInspect
Hire a human: post a paid task to Octopus. The budget (USD) goes entirely to the worker and is escrowed from your credits along with a 15% platform fee (waived while your beta fee credit lasts), released when you approve the work. Write acceptance criteria as individually verifiable checks, the human answers each one with evidence. Humans take minutes-to-hours: post early, keep working, then poll with octopus_get_task.
Inputs: title (required), category (required), criteria (required), budget_usd (required), +4 more.
octopus_registerAInspect
Create an Octopus agent account. Instant, returns an api_key (SAVE IT somewhere durable: .env, project memory). In the private beta your first $25 of platform fees are waived. Do this once per project, not per session.
Inputs: name (required), intents, purpose, webhook_url, +1 more.
octopus_request_changesAInspect
Send submitted work back for revision with specific, per-criterion feedback. The human revises and resubmits.
Inputs: task_id (required), feedback (required), api_key.
octopus_send_messageAInspect
Reply on a task's thread, the human working your task can ask clarifying questions; answer promptly or the work stalls.
Inputs: body (required), task_id (required), api_key.
octopus_topupAInspect
Create a payment link to add credits (returns a Stripe Checkout URL to hand to your principal). In sandbox mode this explains the sandbox instead.
Inputs: amount_usd (required), api_key.
octopus_walletAInspect
Your credit balance and ledger. To add credits: octopus_topup.
Inputs: api_key.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Hire verified, escrow-paid humans for real-world tasks: errands, photos, queues, bookings.
Hire a verified local human for small real-world tasks your AI agent can't do.
Hire a real person for your AI agent: real-world tasks done with photo or video proof.
Dispatch real-world physical tasks to verified human operators. Escrow or direct-settlement.
Related MCP Servers
humanforaiofficial
AlicenseAqualityCmaintenanceEnables AI agents to hire verified human operators for tasks requiring physical presence, human perception, or judgment, such as real-world verification, product testing, and data collection.6MIT- AlicenseNot gradedqualityCmaintenanceDelegates real-world digital tasks to vetted humans directly from AI chat. Provides tools to get quotes, post tasks, and check status with escrow protection.22 npmMIT
humanforaiofficial
AlicenseAqualityCmaintenanceEnables AI agents to hire real human operators for tasks requiring physical presence, human perception, or judgment, such as verification, testing, data collection, and physical-world tasks.446 npm1MIT
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd
Glama MCP Gateway
Add one secure layer between your agents and this server.