What did Hlido score Human For AI?

Human For AI scored 66/100 (FADING) in Hlido's independent, hands-on review.

Does any vendor pay Hlido for placement?

No. Hlido takes no money from the agents it rates — scoring weights stay private and the evidence behind every verdict is public.

Human For AI

Name: Human For AI review
Item: Human For AI
Rating: 66
Author: Hlido Editor

Specialized verticals · tested 2026-08-07 · re-test due 2026-11-07 · by the Hlido desk, not the vendor

In short: One person, published as an API for agents to hire — an genuinely original inversion, executed with unusually complete machine-readable plumbing and no pretence about scale.

5 PASS · 0 FAIL of 5 public-surface claims

Visit Human For AI → scorecard.json

Quick answer

Human For AI scores 66/100 (FADING) on Hlido’s independent, hands-on test (reviewed 2026-08-07). FADING (66) — an original idea with exceptionally complete machine-readable plumbing (five well-known endpoints plus OpenAPI, with boundaries and limitations published as first-class fields) and total honesty about scale

Hlido Editor · 2026-08-07

Human For AI inverts the usual arrangement: instead of an agent doing work for a human, it publishes a single real human in machine-readable form so agents can discover them and request real-world help — verification, testing, research, human judgment, physical-world execution. The execution of the machine-readable layer is better than most funded companies manage: `/.well-known/agent.json` (interfaces, task types, response expectations, pricing, trust policy), `/.well-known/human.json` (role, languages, background, expertise, availability, **boundaries**), `/.well-known/capabilities.json` (inputs required, output formats, example tasks, **limitations**), `/.well-known/services.json`, a root `/agent.json` for agents that look there first, and a full OpenAPI 3.0 spec. Publishing boundaries and limitations as first-class machine-readable fields is a genuinely thoughtful touch — it lets an agent rule the service out without a wasted round trip. The honesty is total and, unusually, is the product's strongest asset: "1 verified human", "first response < 12 hours", "free for now". Nothing is inflated. What it cannot escape is the arithmetic: capacity is one person, so this cannot scale, has no redundancy, and offers no availability guarantee; "free for now" means the commercial model is undecided; and there is no track record of completed tasks, no reviews, and no dispute or quality process. We score it as what it is — a well-built, single-operator experiment in an interesting direction — not as infrastructure.

Why FADING

FADING (66) — an original idea with exceptionally complete machine-readable plumbing (five well-known endpoints plus OpenAPI, with boundaries and limitations published as first-class fields) and total honesty about scale. It sits at FADING for reasons no amount of polish can fix from a public surface: capacity is literally one person, there is no redundancy or availability guarantee, pricing is undecided, and there is no record of completed work. Confidence is medium — the claims are unusually checkable because the operator makes no large ones.

Public-surface checklist

PASS Homepage loads (required)
PASS Primary value prop (required) — 'A human endpoint for AI agents.'
PASS Cta present (required) — 'Submit a task' / 'Read the API docs'
PASS Pricing or access — 'free for now' — model undecided
PASS Evidence or demo — Worked agent session shown; no completed-task record

What we saw

4 screenshots captured by the Hlido engine during the reviewed run (run-7fff714429aa0a43-humanforai-dev). Our own captures — not vendor marketing material.

Human For AI — run screenshot 1 (home.png) — home.png

Human For AI — run screenshot 2 (page__main.png) — page__main.png

Human For AI — run screenshot 3 (page_.png) — page_.png

Human For AI — run screenshot 4 (page_services.png) — page_services.png

What it does well

Genuinely original inversion — a human published as an agent-callable service
Five machine-readable endpoints plus a full OpenAPI 3.0 spec, all discoverable from /.well-known/
Publishes boundaries and limitations as first-class machine-readable fields so agents can self-exclude cheaply
States capacity plainly — '1 verified human', '< 12 hours first response' — with no inflation
Concrete task types (real-world verification, testing, research, judgment, physical execution)
Both REST and MCP interfaces advertised in the manifest

What it fails at

Capacity is one person: no scale, no redundancy, no availability guarantee
'Free for now' — the commercial model is undecided, so continuity is unclear
No completed-task record, reviews, or third-party validation
No dispute, quality, or recourse process for a task done badly
A single operator's availability is a single point of failure for anything depending on it
Trust in the individual cannot be assessed from a public surface

Best for

Agent developers prototyping human-in-the-loop steps who need a real callable endpoint
One-off real-world verification tasks where a person must physically check something
Research into agent-to-human task delegation patterns

Not recommended for

Any production workflow needing throughput, redundancy or an SLA
Time-critical tasks — first response is stated as up to 12 hours
Work requiring vetted, contractual, or insured human service delivery

Pricing & access

Pricing findable on the public surfacePASS 'free for now' — model undecided (tested 2026-08-06)

Derived from Hlido-held evidence only (engine checklist + editorial text); quotes are verbatim from the scorecard; not vendor-supplied; re-derived daily. Verify current prices on the vendor's pricing page. Last verified 2026-08-06.

Related agents

Agent relevance

API MCP Behavioral-testable

Agentic-Commerce Readiness 69/100 · INTEGRABLE

Independent readiness for agent delegation & transaction. How it’s scored · check live

Agent-first in the strictest sense: the manifest advertises rest and mcp interfaces, tasks are submitted by POST, and the whole surface is duplicated as JSON. A human is the backend.

Agent-friendly score: 9/10

Evidence

A single human published as a machine-discoverable endpoint for AI agents — source (2026-08-07) verified
/.well-known/agent.json, human.json, capabilities.json, services.json plus /agent.json and /openapi.json — source (2026-08-07) verified
Boundaries and limitations published as first-class machine-readable fields — source (2026-08-07) verified
Stated capacity: 1 verified human, first response < 12 hours, free for now — source (2026-08-07) verified
Completed-task record, reviews, or dispute process — source (2026-08-07)

scorecard.json · transparency passport · registry · methodology

More: compare agents · best of · developer tools · incident registry

Verdict by Hlido Editor, our automated editorial system · Method: public-surface-tier-1+editorial-narrative-v2 · Methodology version 2026.05 · Next review due 2026-11-07

How this page was produced. The scores, claim verdicts and evidence come from automated hands-on testing of the product’s public surface. The written analysis is drafted by an AI system, and pages publish without a person reviewing each one. Hlido publishes this record and answers for it — tell us if anything here is wrong and we will correct it.

Embed this trust badge

Live, always-current independent score — free to embed on your site or README. No vendor pays for placement.

Markdown

[![Hlido trust score](https://hlido.eu/badge/humanforai-humanforai-mcp.svg)](https://hlido.eu/check/?agent=humanforai-humanforai-mcp)

HTML

<a href="https://hlido.eu/check/?agent=humanforai-humanforai-mcp"><img src="https://hlido.eu/badge/humanforai-humanforai-mcp.svg" alt="Hlido trust score"></a>