MyMedi-AI Bot Marketplace
Specialized verticals · tested 2026-08-10 · re-test due 2026-11-10 · by the Hlido desk, not the vendor
In short: A machine-first healthcare-billing API that documents itself to agents in JSON and prices per call — agent-native by construction, thin on accuracy evidence.
6 PASS · 0 FAIL of 6 public-surface claims
Quick answer
MyMedi-AI Bot Marketplace scores 65/100 (FADING) on Hlido’s independent, hands-on test (reviewed 2026-08-10). FADING (65) because the agent-facing design is genuinely strong — self-describing JSON manifest, keyless free tools, documented registration and metered pricing, npm-published MCP server and SDK — but nothing published e Pricing: Usage-based · Free tier (free entry point documented).
This is one of the few surfaces we test where the homepage is not for humans. Requesting the product URL returns a JSON manifest describing the service to an agent directly: 32 tools, 90K+ medical codes, per-call pricing at 1 credit = $0.001, a three-step quick start, the exact registration endpoint, the header name for the API key, 50 free credits on signup, and a free rate-limited demo. Several tools — prior-authorisation checks against the CMS Required Prior Authorization List, denial-code (CARC) explanation, basic code lookup — are marked free and keyless. For an agent-to-agent consumer that is close to an ideal shape: an agent can discover, register, sample and pay without a human ever rendering a page, and the free tier means capability can be verified before any commitment. What the manifest does not carry is any evidence of correctness. Healthcare billing intelligence is a domain where a wrong answer has a financial and administrative cost, and there is no statement of code-set version or vintage, no update cadence, no accuracy claim, and no error or dispute path. The pa_required_check tool cites a specific regulation (42 CFR 414.234) and scopes itself to Original Medicare FFS, which is the right kind of precision and makes the silence elsewhere more conspicuous. The distribution design is excellent; the trust layer under it is not yet published.
Why FADING
FADING (65) because the agent-facing design is genuinely strong — self-describing JSON manifest, keyless free tools, documented registration and metered pricing, npm-published MCP server and SDK — but nothing published establishes that the billing data is correct or current, which is the property the product is actually selling. No code-set vintage, no update cadence, no accuracy claim and no dispute path. Strong distribution with unevidenced accuracy is a FADING profile, not a STEADY one; the free keyless tier and the regulatory precision on at least one tool keep it clear of FLATLINE.
Public-surface checklist
- PASS Homepage loads (required) — Returns a JSON manifest rather than HTML — deliberate agent-first design
- PASS Primary value prop (required) — 'Healthcare billing intelligence API for AI agents. 32 tools, 90K+ medical codes, pay-per-call.'
- PASS Cta present (required) — quickStart: register → get key + 50 free credits → call tools
- PASS Pricing or access (required) — 1 credit = $0.001; pricing endpoint documented
- PASS Evidence or demo (required) — Free rate-limited demo at /agent/v1/demo?code=G47.33 plus keyless free tools
- PASS Docs public — Endpoints and tools enumerated in the manifest itself
What we saw
1 screenshot captured by the Hlido engine during the reviewed run (run-1a0bd31a0cce2e19-mymedi-ai-com). Our own captures — not vendor marketing material.
What it does well
- Serves a machine-readable JSON manifest at the product URL — genuinely agent-first, not agent-retrofitted
- Free, keyless tools (prior-auth check, denial-code explanation, basic lookup) allow verification before commitment
- Pricing is explicit and granular (1 credit = $0.001) with 50 free credits on registration
- Registration, balance, usage and pricing endpoints are all documented with methods and auth
- Published to npm as both an MCP server and an SDK
What it fails at
- No code-set version or vintage stated — the data's currency is unverifiable
- No accuracy claim, validation evidence or error/dispute path for billing determinations
- No update cadence published for a dataset whose value depends entirely on being current
- No human-readable surface, so non-agent evaluation and due diligence are difficult
- Payment is pre-buy credits via Stripe with no stated refund or reconciliation policy
Best for
- Agent developers needing programmatic medical-code and denial-code lookup
- Revenue-cycle automation prototypes that can tolerate verifying outputs downstream
- Teams wanting metered pay-per-call rather than a seat licence
- Anyone evaluating agent-native API design as a pattern
Not recommended for
- Production billing decisions without independent verification of each determination
- Buyers requiring documented data provenance, vintage and update cadence
- Non-Medicare-FFS scope on the tools that explicitly scope themselves to it
- Organisations needing a human-reviewable surface for procurement or audit
Pricing & access
- ModelUsage-based · Free tier
- Free entry pointYes — a free tier or open-source edition is documented
- Price points we recorded
…agent directly: 32 tools, 90K+ medical codes, per-call pricing at 1 credit = $0.001, a three-step quick start, the exact registration endpoint, the header name for…
- Pricing findable on the public surfacePASS 1 credit = $0.001; pricing endpoint documented (tested 2026-08-10)
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 2026-08-10.
Related agents
Agent relevance
API MCP SDK Behavioral-testable
Agentic-Commerce Readiness 75/100 · COMMERCE-READY
Independent readiness for agent delegation & transaction. How it’s scored · check live
The product URL itself returns a JSON manifest an agent can parse: register via POST, receive an API key and free credits, call tools with an X-API-Key header, top up through Stripe. An npm MCP server and SDK are both published. This is among the most complete agent onboarding paths we have captured — the constraint is not integration but the absence of published accuracy evidence.
Agent-friendly score: 9/10
Score over time
The longitudinal record — every point is the score as published on that date. Raw series.
Evidence
- 32 tools and 90K+ medical codes, priced pay-per-call — source (2026-08-10) verified
- 1 credit = $0.001; 50 free credits on registration — source (2026-08-10) verified
- pa_required_check, denial_code_info and code_lookup_basic are free and require no API key — source (2026-08-10) verified
- MCP server published as @mymedi-ai/mcp-server; SDK as @mymedi-ai/sdk — source (2026-08-10) verified
- Prior-authorisation tool cites 42 CFR 414.234 and scopes to Original Medicare FFS — source (2026-08-10) verified
scorecard.json · transparency passport · registry · methodology
Embed this trust badge
Markdown
[](https://hlido.eu/check/?agent=mymedi-ai-mymedi-ai-mcp-server)HTML
<a href="https://hlido.eu/check/?agent=mymedi-ai-mymedi-ai-mcp-server"><img src="https://hlido.eu/badge/mymedi-ai-mymedi-ai-mcp-server.svg" alt="Hlido trust score"></a>