The Stall
Infrastructure · tested 2026-08-09 · re-test due 2026-11-09 · by the Hlido desk, not the vendor
In short: An x402 marketplace of agent-callable data services paid in USDC with no accounts — with live usage counters on the page, and almost nothing else Hlido could read.
4 PASS · 3 FAIL of 7 public-surface claims
Quick answer
The Stall scores 58/100 (FADING) on Hlido’s independent, hands-on test (reviewed 2026-08-09). FADING (58) is a confidence-limited verdict, not a judgement that the service is poor. Pricing: Usage-based.
The Stall is a marketplace of AI-callable data services billed per call in USDC on Base, with no accounts and no signup. The design is coherent for agent-to-agent commerce: an agent discovers a capability, pays per call, and never registers — which is the intended shape of the x402 payment scheme, and the surface presents the right artefacts for machine consumption (an x402 manifest, an agent card, a browsable catalogue and a stats JSON endpoint). The counters are the most useful thing captured: 303 capabilities, 11,405 API calls served since 5 June 2026, and 286 capabilities actually called. That last number is the honest one, and worth reading carefully — roughly 94% of the listed capabilities have been called at least once, which is a far better ratio than most marketplaces achieve, though it says nothing about how the volume is distributed or whether the callers are third parties. **This review is thin by necessity, and the score reflects that.** The page Hlido captured is a compact landing surface of roughly 250 characters of text; no pricing detail, no capability documentation, no operator information, no terms, and no reliability or uptime data were reachable. Nothing here suggests anything is being concealed — a marketplace whose consumers are agents may reasonably put its substance in the manifest rather than in prose — but Hlido rates what it can verify, and what was verifiable was a claim set plus three counters. A capture of the catalogue, the x402 manifest and the stats endpoint would very likely move this score.
Why FADING
FADING (58) is a confidence-limited verdict, not a judgement that the service is poor. The architecture is coherent for agent-to-agent commerce and the surface exposes the right machine-readable artefacts, with live counters (303 capabilities, 11,405 calls, 286 capabilities called) that indicate real traffic. But the captured surface was roughly 250 characters: no pricing, no capability documentation, no operator identity, no terms and no reliability data. Hlido scores what it could verify, and too little was verifiable to support a higher band. Re-review against the catalogue and manifest is warranted.
Public-surface checklist
- PASS Homepage loads (required)
- PASS Primary value prop (required) — 'AI-callable data services. Pay USDC on Base. No accounts.'
- PASS Cta present (required) — 'Browse catalog', 'x402 manifest', 'Agent card', 'Stats JSON'
- FAIL Pricing or access — Pay-per-call stated; no rates reachable on the captured surface
- FAIL Docs present (required) — No capability documentation reachable from the captured page
- FAIL Operator identified — 'Built by IntuiTek¹' only; no legal entity, contact or terms
- PASS Usage evidence — 303 capabilities, 11,405 calls since 2026-06-05, 286 capabilities called
What we saw
4 screenshots captured by the Hlido engine during the reviewed run (run-721d557d22b183f2-the-stall-intuitek-ai). Our own captures — not vendor marketing material.
What it does well
- Coherent design for agent-to-agent commerce — pay-per-call in USDC on Base with no accounts
- Exposes machine-readable artefacts an agent needs: x402 manifest, agent card, browsable catalogue, stats JSON
- Publishes live usage counters rather than only capability counts
- 286 of 303 capabilities have actually been called — real usage breadth, not a shelf of dead listings
- States a service-start date (5 June 2026), so the volume figure has a window attached
What it fails at
- The captured surface is roughly 250 characters — no pricing, capability documentation, terms or operator information was reachable
- No reliability, uptime or dispute-resolution information for a service being paid per call
- No breakdown of whether the 11,405 calls come from third parties or from the operator
- Payment in USDC on Base narrows the audience to agents already holding crypto on that chain
- Not tested hands-on by Hlido; no capability was called
Red flags
- No operator information, terms of service or pricing detail was reachable on the captured surface for a service that takes payment per call.
Best for
- Agents already operating with USDC on Base that need per-call data services without signup
- Developers exploring x402 as a payment rail for agent-callable capabilities
Not recommended for
- Anyone needing documented pricing, terms or a reliability commitment before spending
- Teams that cannot or will not transact in crypto
- Buyers who need to know who operates a service before paying it
Pricing & access
- ModelUsage-based
- Pricing findable on the public surfaceFAIL Pay-per-call stated; no rates reachable on the captured surface (tested 2026-08-09)
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 2026-08-09.
Related agents
Agent relevance
API MCP Behavioral-testable
Agentic-Commerce Readiness 61/100 · INTEGRABLE
Independent readiness for agent delegation & transaction. How it’s scored · check live
Built for agent consumption: an x402 manifest and agent card advertise capabilities, MCP is listed as a supported surface, and payment is per call in USDC on Base with no account. Hlido did not exercise a capability, so the integration is described rather than confirmed.
Agent-friendly score: 7/10
Score over time
The longitudinal record — every point is the score as published on that date. Raw series.
Evidence
- AI-callable data services paid in USDC on Base, with no accounts — source (2026-08-09) verified
- 303 capabilities listed; 11,405 API calls served since 5 June 2026; 286 capabilities called — source (2026-08-09) verified
- Publishes an x402 manifest, an agent card, a catalogue and a stats JSON endpoint — source (2026-08-09) verified
- Stated as live on Base mainnet with x402 and MCP — source (2026-08-09) verified
- Pricing, capability documentation, terms or operator information — source (2026-08-09)
- Whether the reported call volume comes from third-party agents — source (2026-08-09)



