{
  "markdown": "# @pipeworx/medicaid-intelligence\n\nMedicaid drug utilization, managed-care enrollment and state enrollment operations from the\nCMS Medicaid Open Data API. Keyless.\n\nPart of [Pipeworx](https://pipeworx.io) — an MCP gateway connecting AI agents to 1476+ live data sources.\n\n## Tools\n\n| Tool | What it returns |\n|---|---|\n| `medicaid_drug_utilization` | Quarterly state utilization rows for one NDC, fee-for-service and managed care kept separate |\n| `medicaid_drug_state_market` | One NDC aggregated across states for a year |\n| `medicaid_drug_trend` | Annual prescriptions, units and gross reimbursement for one NDC from 2020 |\n| `medicaid_managed_care_summary` | Annual state enrollment, and enrollment in any or comprehensive managed care |\n| `medicaid_managed_care_program_mix` | State enrollment by program type — comprehensive MCO, PCCM, MLTSS, behavioral health, dental, transportation, PACE |\n| `medicaid_plan_market` | Bounded sample of managed-care plan/program rows by state, with the authoritative matching-row count |\n| `medicaid_monthly_managed_care` | Monthly Medicaid/CHIP enrollment for one state and participation category |\n| `medicaid_enrollment_operations` | Monthly enrollment and application-processing indicators for one state |\n\n## Auth\n\nNone. Every endpoint is public CMS infrastructure at `data.medicaid.gov`.\n\n## Reading these numbers correctly\n\nThis source is easy to misread, and most of the pack's design is about preventing that. Each\nresponse carries an `interpretation` field stating the caveat that applies to it.\n\n- **Reimbursement is gross, before rebates.** `total_amount_reimbursed` is what state Medicaid\n  programs paid pharmacies. It is neither manufacturer revenue nor net price — Medicaid rebates\n  are substantial and are not in this data. Treating these figures as a manufacturer's Medicaid\n  revenue overstates it, often by a lot.\n- **Suppressed is not zero.** CMS suppresses small cells (`DS`, `*`, `--`, or a\n  `suppression_used` flag). Those become `null`, never `0`, and a channel that is entirely\n  suppressed is reported as unavailable rather than as an absence of utilization. Tests guard\n  this specifically.\n- **Fee-for-service and managed care are separate measures** and stay separate. Adding them is\n  not meaningful for most questions, because a state's mix shifts over time.\n- **Managed-care program categories overlap and must not be summed.** A beneficiary can appear\n  under comprehensive MCO and MLTSS and behavioral health. `medicaid_managed_care_program_mix`\n  says so in its `interpretation`.\n- **Enrollment counts are state-reported program enrollment**, not covered lives attributable to\n  a particular insurer. They do not map cleanly onto a payer's book of business.\n- **A zero in the plan file can be confidentiality suppression**, not an empty plan.\n  `medicaid_plan_market` returns a bounded sample plus the authoritative total row count, so a\n  short list never implies a small market.\n\n## Argument conventions\n\nThe underlying datasets disagree about how a state is written — the drug files use two-letter\ncodes, the enrollment and managed-care files use full names. **Both forms are accepted on every\ntool** and converted internally. This matters: before that, asking the managed-care summary for\n`\"CA\"` matched nothing and returned a confident `total: 0`, which reads as \"California has no\nmanaged care\" rather than as a wrong argument.\n\n`ndc` takes an exact 11-digit National Drug Code, hyphenated or not. A brand or ingredient name\nreturns `{found:false, reason:'ndc_expected_got_name'}` with a hint pointing at `openfda_drug_label`\nor `rxnorm` to resolve it first.\n\n## Failure shapes\n\nA recoverable caller mistake resolves to `{found:false, reason, hint}` naming what to do instead.\nAn upstream failure or a genuine defect resolves to `{error, hint}`. The two are deliberately\ndistinguishable, so a monitoring sweep can tell \"CMS refused us\" apart from \"we returned garbage\".\n\n## Data sources\n\nAll datasets are queried through the CMS Medicaid Open Data datastore API,\n`https://data.medicaid.gov/api/1/datastore/query`:\n\n- [State Drug Utilization Data](https://data.medicaid.gov/dataset?tags=State%20Drug%20Utilization%20Data) — one dataset per year, 2020 through 2026\n- [Managed Care Enrollment Summary](https://data.medicaid.gov/dataset/52ed908b-0cb8-5dd2-846d-99d4af12b369) and [enrollment by program](https://data.medicaid.gov/dataset/e2ce0d2f-07c5-5213-947a-31e19bc649f6)\n- [Managed Care Plan and Program data](https://data.medicaid.gov/dataset/0bef7b8a-c663-5b14-9a46-0b5c2b86b0fe)\n- [Monthly Medicaid and CHIP managed-care enrollment](https://data.medicaid.gov/dataset/89baf100-259b-4763-b9e2-337972f988c4)\n- [Medicaid and CHIP enrollment and operations reports](https://data.medicaid.gov/dataset/6165f45b-ca93-5bb5-9d06-db29c692a360)\n\n## Quick Start\n\nAdd to your MCP client (Claude Desktop, Cursor, Windsurf, etc.):\n\n```json\n{\n  \"mcpServers\": {\n    \"medicaid-intelligence\": {\n      \"url\": \"https://gateway.pipeworx.io/medicaid-intelligence/mcp\"\n    }\n  }\n}\n```\n\n### What this endpoint actually serves\n\n`tools/list` at `https://gateway.pipeworx.io/medicaid-intelligence/mcp` returns the tools in the table\nabove **plus the shared Pipeworx meta-tools** — `ask_pipeworx`,\n`discover_tools`, `search_within`, `remember`/`recall` and the rest of the\ngateway-wide set. So the tool count you see is larger than this table: a\nsingle-pack endpoint currently lists roughly 30 shared tools alongside the\npack's own. The connection's `initialize` response states its exact scope, and\nis the authoritative answer for a given day.\n\nThis is deliberate, not multiplexing by accident. The meta-tools are what let a\nscoped connection answer a question this pack does not cover — via\n`ask_pipeworx`, which routes across the whole catalog — without you adding a\nsecond MCP server. There is currently no way to mount a pack endpoint without\nthem; if the extra schemas cost you more context than the routing is worth,\nconnect to the full gateway once rather than to several pack endpoints.\n\nOr connect to the full Pipeworx gateway to get every pack's tools listed\ndirectly, instead of just this one's:\n\n```json\n{\n  \"mcpServers\": {\n    \"pipeworx\": {\n      \"url\": \"https://gateway.pipeworx.io/mcp\"\n    }\n  }\n}\n```\n\nBoth URLs reach the same gateway and the same 1476+ data sources. The\nonly difference is which pack's tools are listed **directly**; `ask_pipeworx`\nreaches all of them from either one.\n\n## Using with ask_pipeworx\n\nInstead of calling tools directly, you can ask questions in plain English —\nthis works on the pack endpoint above as well as on the full gateway:\n\n```\nask_pipeworx({ question: \"your question about Medicaid Intelligence data\" })\n```\n\nThe gateway picks the right tool and fills the arguments automatically.\n\n## More\n\n- [Docs and guides](https://pipeworx.io/docs)\n- [pipeworx.io](https://pipeworx.io)\n\n## License\n\nMIT\n",
  "bytes": 6906,
  "sha": "e981a02c7ef88009aea86b37032166415d7807be1fc7cae6e3b00bf1c95103aa",
  "repo_slug": "pipeworx-io/mcp-medicaid-intelligence",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_pipeworx_io_medicaid_intellige_e065b1ba/readme"
}