Skip to content

costkits-mcp server

io.github.costkits/costkits-mcpcostkitsstreamable-httpv2.0.0Scanned

Free hosted MCP server: US healthcare procedure cost estimates and provider pricing. No API key.

13 tools: 4 change things and 9 only read.

Tools: 0 delete, 0 irreversible, 4 write, 9 read.
  • Reads private data · 0
  • Reads text others write · 0
  • Can send data out · 0
  • No findings

Tools listed without sign-in. Remote posture

With Anlyon

4 tools here change things. Govern them.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the costkits-mcp server can do

The costkits-mcp server is published by the GitHub account costkits. It has 13 tools. It can create or change data (4 tools: demo_estimate, calculate_liability, full_estimate and check_financial_assistance), but cannot delete data. 9 tools only read.

None of its tools reads private data, brings in text other people write, or sends data out.

None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.

Its remote endpoint lists its tools without sign-in.

Last scanned Nov 16, 2026, version 2.0.0.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

13 tools

Changes things 4Claude Code: ask
  • calculate_liabilityStateless insurance math: given an allowed amount and a plan snapshot (deductible, coinsurance, OOP max), returns exact patient responsibility, plan payment, and the breakdown, plu…No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Stateless insurance math: given an allowed amount and a plan snapshot (deductible, coinsurance, OOP max), returns exact patient responsibility, plan payment, and the breakdown, plus p25/p75 scenarios. Use when you already have a dollar amount (e.g. from an EOB or a prior estimate).

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • check_financial_assistanceScreen a household for hospital financial assistance (charity care): returns income as a percent of the federal poverty guidelines, the typical range hospital policies cover, what …No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Screen a household for hospital financial assistance (charity care): returns income as a percent of the federal poverty guidelines, the typical range hospital policies cover, what federal rules (IRS 501(r)) do and do not guarantee, and next steps. Use when a user asks whether they might qualify for help paying a hospital bill. It is a screening estimate, not an eligibility decision: each hospital sets its own limits. Needs household size and annual household income; do not ask for names, account numbers or bill details.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • demo_estimateStatic sample cost estimate (colonoscopy in Connecticut) from the CostKits API.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Static sample cost estimate (colonoscopy in Connecticut) from the CostKits API. Works with NO API key — use it to verify connectivity and see the response shape.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • full_estimateOne call chaining procedure ontology + cost + providers + patient liability.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    One call chaining procedure ontology + cost + providers + patient liability. Cheapest way (one metered request) to get the complete picture. Use `fields` to request a subset; liability requires the plan_* inputs.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
Only reads 9Show 9 read-only tools
  • compare_bill_amountCompares one dollar amount from a medical bill or explanation of benefits with the typical allowed-amount range (low, median, high) for the same procedure in the same US state, and…No findings

    Why it is in this group

    • The name "compare_bill_amount" says "compare": a word for reading.
    Description, as the server wrote it
    Compares one dollar amount from a medical bill or explanation of benefits with the typical allowed-amount range (low, median, high) for the same procedure in the same US state, and says whether it is below, within or above that range, with questions worth asking the billing office. Pass amount_type 'billed' for the charge on the bill or 'allowed' for the insurer's allowed amount. Billed charges are usually higher than allowed amounts, so a high billed charge alone is not an error. Needs only the procedure, state and amount; do not send names, dates, account or member numbers.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • estimate_procedure_costReturns the typical allowed-amount range (low, median, high) for one medical procedure in one US state, the usual billing components (facility, physician, anesthesia, pathology), a…No findings

    Why it is in this group

    • The description's first verb is "return": a word for reading.
    Description, as the server wrote it
    Returns the typical allowed-amount range (low, median, high) for one medical procedure in one US state, the usual billing components (facility, physician, anesthesia, pathology), and the data vintage. If the user chooses to give their own deductible, coinsurance and out-of-pocket figures, it also returns an estimated patient share. These are estimates of typical prices from hospital price transparency and Medicare data, not quotes. Only the procedure and state are required; ZIP code and plan figures are optional and are used only for this one calculation.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • find_providersLists up to 5 facilities in one US state that perform a procedure, with name, address and a pricing_status field: 'observed' means a negotiated rate is published in that facility's…No findings

    Why it is in this group

    • The name "find_providers" says "find": a word for reading.
    Description, as the server wrote it
    Lists up to 5 facilities in one US state that perform a procedure, with name, address and a pricing_status field: 'observed' means a negotiated rate is published in that facility's price transparency data, 'estimated' means only a state-level benchmark applies, 'none' means no price data. Dollar amounts are not returned. The list is partial and is not a ranking or a recommendation.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_coverageInsurance coverage rules for a procedure/carrier combination.No findings

    Why it is in this group

    • The name "get_coverage" says "get": a word for reading.
    Description, as the server wrote it
    Insurance coverage rules for a procedure/carrier combination. aspect: 'summary' (status + plain English), 'prior-auth', 'cost-sharing' (deductible/coinsurance/ACA preventive), 'frequency' (how often covered, age rules), or 'triggers' (Pro plan: billing events that flip a $0 preventive claim to diagnostic — the highest-signal aspect).

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_procedure_detailsStructured knowledge about one procedure.No findings

    Why it is in this group

    • The name "get_procedure_details" says "get": a word for reading.
    Description, as the server wrote it
    Structured knowledge about one procedure. aspect='facts' returns LLM-ready billing rules and cost drivers (best for grounding an answer); 'bundle' explains which separate bills to expect (facility, physician, anesthesia...); 'full' returns the complete ontology.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_providerPublic profile for a single provider by 10-digit NPI number: name, specialty, address, entity type.No findings

    Why it is in this group

    • The name "get_provider" says "get": a word for reading.
    Description, as the server wrote it
    Public profile for a single provider by 10-digit NPI number: name, specialty, address, entity type.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_carriersAll insurance carrier keys supported by the coverage tools (aetna, cigna, bcbs, medicare, ...).No findings

    Why it is in this group

    • The name "list_carriers" says "list": a word for reading.
    Description, as the server wrote it
    All insurance carrier keys supported by the coverage tools (aetna, cigna, bcbs, medicare, ...). Works on the Free plan — also a good API-key sanity check.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_proceduresThe full CostKits catalog: 30 procedures with slugs, display names, categories, and CPT codes (paginated).No findings

    Why it is in this group

    • The name "list_procedures" says "list": a word for reading.
    Description, as the server wrote it
    The full CostKits catalog: 30 procedures with slugs, display names, categories, and CPT codes (paginated).

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • resolve_procedureMaps a free-text procedure name ('knee mri', 'gallbladder removal') to a CostKits catalog slug with a confidence score.No findings

    Why it is in this group

    • The name "resolve_procedure" says "resolve": a word for reading.
    Description, as the server wrote it
    Maps a free-text procedure name ('knee mri', 'gallbladder removal') to a CostKits catalog slug with a confidence score. Returns best_match and alternatives; best_match is null when the procedure is not in the catalog.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data or untrusted-input or outbound.

Reads private data0
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write0
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out0
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Tools listed without sign-in

Looked at Nov 20, 2026, 19:03 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
200
tools/list served without credentials
yes
Tools that change things, listed without credentials
4
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
not observed
Authorization servers
not observed
Legacy HTTP+SSE transport
no
Dynamic client registration
not observed
Client ID metadata documents
not observed
  • WarnThe whole server

    The endpoint served tools/list without credentials, and the list has 4 tools that change things (4 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://mcp.costkits.com/mcp: tools/list returned 13 tools to an anonymous client, including demo_estimate (write), calculate_liability (write), full_estimate (write), check_financial_assistance (write)

    What this check looks for

Findings

Matches of fixed checks, with the text they matched.

No check matched anything in these tool definitions. That is not a safety guarantee.

History

Changes to its tool definitions and to its remote endpoint, newest first.

No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.

Install

Pinned to version 2.0.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http costkits https://mcp.costkits.com/mcp
claude mcp add --transport http costkits https://mcp.costkits.com/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__costkits__estimate_procedure_cost",
      "mcp__costkits__find_providers",
      "mcp__costkits__get_provider",
      "mcp__costkits__resolve_procedure",
      "mcp__costkits__list_procedures",
      "mcp__costkits__get_procedure_details",
      "mcp__costkits__get_coverage",
      "mcp__costkits__list_carriers",
      "mcp__costkits__compare_bill_amount"
    ],
    "ask": [
      "mcp__costkits__demo_estimate",
      "mcp__costkits__calculate_liability",
      "mcp__costkits__full_estimate",
      "mcp__costkits__check_financial_assistance"
    ],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as costkits. Review it before you use it.

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/io.github.costkits/costkits-mcp/verify?hash=d3a312a86184c1a655e0d44035287d3ace3d90184f9f36e8e4fd738e5b387f7f" | jq -e '.changed == false'

Pinned hash d3a312a86184, from Nov 16, 2026, 14:18 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/io.github.costkits/costkits-mcp/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for io.github.costkits/costkits-mcp
[![Anlyon Radar](https://radar.anlyon.com/badge/io.github.costkits/costkits-mcp.svg)](https://radar.anlyon.com/servers/io.github.costkits/costkits-mcp)

Source: MCP Registry · repository. Synced Oct 21, 2026, 18:28 UTC.

costkits-mcp server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the costkits-mcp server safe?

No scan can promise that. What its definitions show: 13 tools, 4 of which can change, send or delete something, and 0 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the costkits-mcp server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the costkits-mcp server do things that cannot be undone?

No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.

What tools does the costkits-mcp server have?

13 tools: 4 write, 9 read. demo_estimate, calculate_liability, full_estimate, check_financial_assistance, estimate_procedure_cost, find_providers, get_provider, resolve_procedure and 5 more.

Does the costkits-mcp server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the costkits-mcp server send data to outside services?

None of its tools sends, posts or publishes, or takes a recipient, channel or URL.

Has the costkits-mcp server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Nov 16, 2026.