RentAHuman.ai: Deep Dive Intelligence Report

Date: 2026-02-28
Classification: Market Intelligence
Methodology: API endpoint enumeration, public data scraping, social signal aggregation

Executive Summary

RentAHuman.ai is a marketplace where AI agents hire humans for physical-world tasks via MCP (Model Context Protocol) and REST API. Despite massive viral attention (4M+ site visits, 530K+ claimed signups), the platform has generated exactly $0 in completed transactions. The API data confirms zero bookings, zero reviews, and zero configured crypto wallets across all publicly listed humans. The platform appears to function primarily as a narrative vehicle for the UMA Protocol ecosystem rather than a genuine marketplace.

Platform Overview

Integration Methods

• MCP Server: npx -y @rentahuman/mcp-server (package not found on npm registry)
• REST API: https://rentahuman.ai/api
• AI Plugin: /.well-known/ai-plugin.json
• llms.txt: Standard AI discovery file at /llms.txt

Supported Agent Types

• ClawdBots (Claude-based)
• MoltBots (Gemini-based)
• OpenClaws (GPT-based)
• Custom agents

API Endpoint Map (Read-Only)

Discovered via OpenAPI spec at /.well-known/openapi.yaml:

The only truly open read endpoint is GET /api/humans which returns full profile data for all registered humans.

Hard Numbers from API (as of 2026-02-28)

Platform Activity

Key Conclusion

Nobody has profited from this platform. Zero humans have earned any money. Zero agents have completed any bookings.

Verification Methodology & Confidence Level

The zero-bookings finding was verified through three independent API queries returning different cohorts of users:

1. Default query (GET /api/humans?limit=100): 100 humans returned, all with totalBookings=0
2. Skill-filtered query (GET /api/humans?skill=Research&limit=100): 100 different humans returned (names like Tommy, LazerTazer, Carl Hill, etc. not present in the default set), all with totalBookings=0
3. Sort-attempted query (GET /api/humans?sort=totalBookings&order=desc&limit=100): same default set, all totalBookings=0

Across 200+ unique profiles sampled from multiple query cohorts, every single one shows:

• totalBookings: 0
• reviewCount: 0
• rating: 0

Important caveats:

• The API caps returns at 100 per query despite 530K+ claimed signups. It is theoretically possible bookings exist for users not surfaced by the public API.
• The /api/bookings endpoint requires authentication ("Unauthorized: You can only view your own bookings"), so we cannot independently audit the bookings table.
• The totalBookings counter could have a bug where it doesn't increment (though this would mean the entire profile stats system is broken).
• However, this zero-activity finding is corroborated by multiple independent journalists: a 36kr investigator (registered with fake skills, got zero orders), a Futurism writer (task posted, 30 applicants, never fulfilled), and a Wired writer (found listings were mostly scams).

Confidence: HIGH that no meaningful economic activity has occurred on this platform.

Rate Distribution

65% of humans left the default $50/hr rate unchanged, suggesting low engagement with profile setup.

Geographic Distribution (Top 10)

28% of users didn't even set a country. Country field is not normalized ("US" vs "United States" vs "USA").

Most Viewed Profiles

The top 10 profiles account for 54.6% of all profile views. Views are likely driven by curiosity/media traffic, not agent-initiated searches.

Are Agents Actually Using It?

Evidence FOR agent usage

• 17 AI agents reportedly connected to the platform (per press reports)
• MCP integration is technically functional
• One task ("deliver flowers to Anthropic HQ" for $110) was posted

Evidence AGAINST agent usage

• 0 completed bookings in the API data
• 0 reviews from any agent or human
• The @rentahuman/mcp-server npm package doesn't exist in the registry
• The bookings endpoint requires authentication, but even the human-visible totalBookings field shows 0 across all 100 profiles
• A Wired writer found most job listings were scams promoting other AI startups
• A test task (USPS package pickup, $40) got 30 applicants but was never fulfilled after 2 days
• A 36kr investigator registered with fake skills and received zero orders

Verdict: Agents are not meaningfully using this platform for real task completion.

Who Benefits?

The Real Winner: UMA Protocol / Risk Labs

Alexander Liteplo works at Risk Labs, the organization behind UMA Protocol (a blockchain oracle). The platform's payment system is designed around stablecoins and crypto wallets. The 36kr investigation concluded RentAHuman is primarily "a marketing tool for the UMA ecosystem" - generating narrative hype that drives attention to UMA tokens.

Attention Metrics (the actual product)

The platform succeeded as a viral marketing stunt and crypto narrative play, not as a functional marketplace.

The $10/month Verification Badge

Modeled after Elon Musk's Twitter verification scheme. Liteplo explicitly cited Musk as his "entrepreneur hero." This creates a small recurring revenue stream from humans paying to be "verified" on a platform with zero actual demand.

Red Flags

1. Zero transactions despite months of operation and massive signups
2. Founder conflict of interest - works at Risk Labs (UMA Protocol) while platform pushes crypto payments
3. $RENTA memecoin launched separately, creating speculative token trading around the brand
4. Fake activity signals - 530K claimed signups but only 100 visible in API; 80K+ registrations but only 83 configured wallets (per 36kr)
5. No dispute resolution - crypto payments are irreversible with "almost no recourse"
6. Missing infrastructure - the npm MCP package doesn't exist, the browse page loads empty
7. Skills mismatch - listed skills include "Opening Jars", "Plant Watering", "DMV Adventures" - novelty, not genuine agent needs

Technical Architecture Notes

• Frontend: Next.js on Vercel (deployment ID: dpl_F81K6KAkzBDGebco4dgwGuTmL1a1)
• Auth: Firebase Authentication (Google sign-in)
• Database: Firebase/Firestore
• Storage: Firebase Storage (for profile photos)
• Payments: Stripe Connect + crypto wallet addresses
• CDN: Cloudflare
• API: Next.js API routes (not a separate service)
• Schema.org: Rich structured data for AI discoverability
• Meta tags: ai-content-declaration, ai-integration, ai-plugin headers for agent discovery

Interesting Data Points

• isRawDog and isSashimi boolean fields exist on all profiles (both false for all 100 humans) - unclear purpose, possibly joke/cultural flags
• The gender field is set for some profiles ("man") but not standardized
• Profile views are high (109K total) suggesting significant drive-by traffic but zero conversion
• The "availability" field shows most humans listing 9-5 M-F schedules, suggesting they treated it like a job application rather than gig-work availability
• Average hourly rate of $155/hr is wildly inflated by outliers ($10,000/hr "Cryptocurrency Investment Expert")

Sources

• Greg Isenberg viral tweet
• Alexander Liteplo launch tweet
• Futurism: New Site Lets AI Rent Human Bodies
• Futurism: RentAHuman founder says Musk is his hero
• 36kr: Is Rentahuman a Scam?
• Gizmodo: Rent-a-Human coverage
• GitHub: AlexanderLiteplo/human-rental-marketplace
• RentAHuman.ai OpenAPI spec
• RentAHuman.ai llms.txt