Skip to main content

gsa-perdiem-mcp

price: free license: MIT tools: 7 regression tests: 576

Free, open-source MCP server for the GSA Per Diem Rates API. Federal travel lodging and M&IE rates for IGCEs and travel cost estimation.

ZIP, state, and M&IE lookups for FY2021 onward work with no key and no network call: they are answered from GSA's published rate, ZIP, and M&IE files bundled in the package. For bundled fiscal years, city lookups, trip estimates and comparisons with a supplied county also work without a key or network call. City-name lookups without a county and lookups for unbundled years use the GSA Per Diem API and need a free api.data.gov key in PERDIEM_API_KEY. The hosted version needs no key for any lookup. See Local or hosted.

Tested and hardened through ten rounds of integration testing against the live GSA Per Diem API, including a round-7 independent re-audit with live verification. 576 collected Python regression tests (309 offline, 267 live-gated) covering 1 P0 path-traversal bug, 23 P1 silent-wrong-data bugs, 21 P2 validation gaps, 14 round-7 findings, the 1.1.0 rate-area ambiguity fixes, and the 1.2.1 and 1.2.2 content-test corrections. See testing.md for the full testing record.

Local or hosted

Local (desktop) Claude (hosted) ChatGPT (hosted)
Local setup Install Install
  • Local runs it on your computer, inside the Claude or ChatGPT desktop app or another AI app. You get your own full rate limits, and it relies only on the government service. Bundled ZIP, state, M&IE and city-plus-county requests need no key. Only live API requests need a free api.data.gov key. You don't have to set it up by hand: give your AI this page's link and ask it to set it up or walk you through it.
  • Hosted is the convenient option: one click, no keys, and it works in Claude or ChatGPT anywhere.

Compare local and hosted · Matching prompts

What it does

Exposes GSA per diem rates as 7 MCP tools:

Core lookups

  • lookup_city_perdiem - Rates by city/state, resolved through the GSA API (optional county answers from GSA's county definitions without an API call)
  • lookup_zip_perdiem - Rates by ZIP code from GSA's ZIP file (optional county)
  • lookup_state_rates - All NSA rates and the standard rate for a state
  • get_mie_breakdown - M&IE tier table (meal components)
  • get_data_status - Bundled fiscal years, their GSA source files, and which tools call the API

Workflow

  • estimate_travel_cost - Calculate trip per diem (lodging + M&IE with first/last day at 75%)
  • compare_locations - Compare rates across multiple cities

Rate areas are never guessed. GSA sets rates by county or locality. When a ZIP or city spans more than one rate area with different rates, results have status: "ambiguous" and list every candidate; supplying county selects one. A city GSA does not recognize is unresolved rather than defaulting to the standard rate. When GSA lists several rate areas for a city, Census 2020 place-to-county records break the tie only if they place the city in exactly one of them. Every result names its source: the bundled GSA file (with SHA-256) or the API endpoint.

Bundled data. src/gsa_perdiem_mcp/data/ holds FY2021 onward, generated by scripts/build_snapshot.py from GSA's per diem files page and Census 2020 place/county code files. data/manifest.json records each source URL and SHA-256. The builder fails on schema drift, on any disagreement between GSA's rate and ZIP files, and on locality wording it has not been taught. FY2020 and fiscal years not yet bundled use the API.

API key (only for live API requests)

ZIP, state and M&IE requests for bundled fiscal years work without a key. So do lookup_city_perdiem, estimate_travel_cost and compare_locations when you supply a county for the bundled year. For example, ask for Boston, MA, in Suffolk County, or include county: "Suffolk" in each Boston comparison entry.

City-name requests without a county, and lookups for unbundled years, call api.gsa.gov and require an api.data.gov key: 1,000 requests per hour, yours alone. Without PERDIEM_API_KEY, those API requests return setup instructions. You can instead supply the work-location county for a bundled year to complete a city lookup or travel estimate without setting up a key.

Get a free key (takes 30 seconds):

  1. Go to api.data.gov/signup
  2. Enter your name and email: no approval, no wait
  3. Copy the key from the confirmation page
  4. Paste it into your client config as PERDIEM_API_KEY (see below)

The same key works for every api.data.gov-backed API (GSA Per Diem, NASA, FEC, FCC, etc.).

Installation

uvx gsa-perdiem-mcp

Configuration

Use the configuration below as the server definition and adapt its placement to your compatible MCP client. For practical requests using this source, see the prompt library.

With your key:

{
  "mcpServers": {
    "gsa-perdiem": {
      "command": "uvx",
      "args": ["--refresh-package", "gsa-perdiem-mcp", "--from", "gsa-perdiem-mcp", "gsa-perdiem-mcp"],
      "env": {
        "PERDIEM_API_KEY": "paste-your-api-data-gov-key-here",
        "FEDERAL_API_MIN_INTERVAL_SECONDS": "4"
      }
    }
  }
}

The server defaults FEDERAL_API_MIN_INTERVAL_SECONDS to 4 for keyed requests. The explicit value above documents the intended policy and can be changed when you have a documented reason. Per Diem and Regulations.gov processes using the same api.data.gov key share the gate.

No-key setup (bundled ZIP, state, M&IE, and city-plus-county lookups, estimates and comparisons work):

{
  "mcpServers": {
    "gsa-perdiem": {
      "command": "uvx",
      "args": ["--refresh-package", "gsa-perdiem-mcp", "--from", "gsa-perdiem-mcp", "gsa-perdiem-mcp"]
    }
  }
}

The --refresh-package flag tells uv to check PyPI for a newer release each time your client launches the server, so fixes arrive automatically; without it, uv keeps serving whatever version it first cached. It adds a moment of network time at startup, so raise your platform's MCP startup timeout if it enforces a short one.

Restart the client and the tools appear.

Example prompts

  • "What's the per diem rate for Washington DC in FY2026?"
  • "Estimate travel costs for 4 nights in Boston in March."
  • "Compare per diem rates for DC, New York, and San Francisco."
  • "What are all the NSA per diem locations in Virginia?"
  • "Show me the M&IE meal breakdown for the $92 tier."
  • "Build a travel estimate: 3 trips to Seattle (4 nights each) and 2 trips to DC (3 nights each)."

Important: maximum reimbursement, not actual prices

Per diem rates are federal reimbursement ceilings per 41 CFR 301-11. They are not actual hotel prices. CONUS only. Non-foreign OCONUS rates (Alaska, Hawaii, territories) are set by DoD (DTMO); foreign rates by the State Department. Lodging taxes generally not included. First/last travel day M&IE at 75%.

Companion tools

Use alongside bls-oews-mcp (wage data) and gsa-calc-mcp (ceiling rates) for complete IGCE development. Per diem covers the travel component; BLS and CALC+ cover labor.

License

MIT

Trips crossing months or fiscal years. Add each month's lodging_total at the applicable fiscal year rate. Calculate M&IE once across the actual travel days: 75% on the departure and return days, 100% on intervening days. Do not add separate monthly mie_total or grand_total values because each estimate assumes its own departure and return. For example, four nights at a $92 daily M&IE rate cost $414 M&IE over five travel days; two independent two-night estimates sum to $460 M&IE.

Metadata

Release files for gsa-perdiem-mcp 1.2.3

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for gsa-perdiem-mcp 1.2.3
File Size Uploaded
gsa_perdiem_mcp-1.2.3.tar.gz 1.3 MB Details

Built distribution (wheel)

Table of built distributions (wheels) for gsa-perdiem-mcp 1.2.3
File Interpreter ABI Platform
gsa_perdiem_mcp-1.2.3-py3-none-any.whl Python 3 none any Details

Total release size: 2.4 MB

Release files / gsa_perdiem_mcp-1.2.3.tar.gz

Download URL gsa_perdiem_mcp-1.2.3.tar.gz
Size 1.3 MB
Tags Source
SHA-256 checksum
How to use checksums
4387a8950f0d4154d35c6150b0809a5c7d81e091d7ad1038af43e181eeeb594a
BLAKE2b-256 checksum
How to use checksums
0c3822bd7f674b4461fd879097f7a1f3a1574ceb3d5b6df068210d216a9a68f7
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 10, 2026.

Transparency log

Release files / gsa_perdiem_mcp-1.2.3-py3-none-any.whl

Download URL gsa_perdiem_mcp-1.2.3-py3-none-any.whl
Size 1.2 MB
Tags Python 3
SHA-256 checksum
How to use checksums
567321d3cbc9c43ec885ff9c17ee9f34eb6b5e4099e5c14f610f5050429eb4ba
BLAKE2b-256 checksum
How to use checksums
bd3833a92256d5d224a1323bc3a5e19948a5b85535bed96de79b6b3f0e1d0926
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 10, 2026.

Transparency log

Release history Release notifications | RSS feed

1.2.6

2 release files

1.2.5

2 release files

1.2.4

2 release files

This release

1.2.3 This release

2 release files

1.2.2

2 release files

1.2.1

2 release files

1.2.0

2 release files

1.1.1

2 release files

1.1.0

2 release files

1.0.11

2 release files

1.0.10

2 release files

1.0.9

2 release files

1.0.8

2 release files

1.0.7

2 release files

1.0.6

2 release files

1.0.5

2 release files

1.0.4

2 release files

1.0.3

2 release files

1.0.2

2 release files

1.0.1

2 release files

1.0.0

2 release files

0.2.6

2 release files

0.2.5

2 release files

0.2.4

2 release files

0.2.3

2 release files

0.2.2

2 release files

0.2.1

2 release files

0.2.0

2 release files

0.1.1

2 release files

0.1.0

2 release files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page