BLS Labor MCP Server

Source-reviewed, not functionally tested

These details were checked against primary sources. Skill Road has not performed a practical client functionality test. Features and setup reflect project documentation, not independently confirmed operation.

US labor statistics with a clear chain of provenance

BLS Labor MCP Server is a community project maintained by cyanheads that exposes US Bureau of Labor Statistics data to MCP clients. BLS supplies the statistics, but the connector itself is not an official agency service. Keep that distinction visible when assessing the project or citing an answer. It is intended for people researching labor and price developments who need a traceable path from a question to a statistical series rather than an unverified number recalled by a language model.

Selecting the right series comes before interpreting a trend

A question about unemployment or inflation can refer to several different measures. Establish the population, geographic scope and observation frequency before comparing results. Decide whether seasonally adjusted or unadjusted data fits the question. For wages, specify whether the analysis concerns an average, a distribution or a particular occupation. Without this preparation, a technically successful request can support the wrong economic claim. Asking for a concise explanation of the chosen series is a useful checkpoint before the assistant constructs a chart or narrative.

The project documents tools for discovering series and retrieving observations. The BLS API associates values with periods and may include explanatory footnotes. Preserve series identifiers, units and date ranges in your work rather than keeping only the assistant's final paragraph. Missing observations should not become zeros, and preliminary values should not be described as settled results. When a summary seems surprising, return to the original statistical source and inspect the selected measure before assuming the model's interpretation is correct.

Local operation and hosted access have different responsibilities

The repository describes local stdio operation and a public community-hosted HTTP endpoint. A BLS API key is optional for basic local use; registration can provide additional quotas. This is not unlimited access. Plan repeated queries within reasonable boundaries and confirm current usage rules with BLS. The hosted connection avoids installation but introduces another operator into the request path. Self-hosting gives you deployment control while also making updates, configuration and access protection your responsibility rather than removing those requirements.

Know what is stored during an analysis

Configuration includes a local catalog cache and optional support for larger staged tabular results. If you enable these facilities, identify their storage location, lifecycle and access boundaries. Dropping DataCanvas tables, disabled by default, is not the same as modifying BLS statistics. Nevertheless, each optional tool expands what the assistant can do in your environment. Enable only what a specific analysis needs and avoid unnecessary privileges on a machine or service shared with other users. Public-source data can still reveal sensitive research interests through query patterns.

Suitable research does not equal automatic decision-making

The Apache-2.0-licensed connector can simplify exploratory work but does not make economic decisions for its users. Reports should document source, data vintage and methodological caveats. General US time series alone are especially insufficient for decisions about individual employees. Do not send personal workforce information to a public host. The integration is most useful for reproducible research on public statistics, with the assistant helping navigate the available material and a person checking comparability, limitations and the interpretation presented to readers.

Source review and verification limits

This entry was reviewed against the linked primary sources on 2 October 2026. Features and setup instructions reflect the project's documentation rather than a complete functional test performed by Skill Road. Before production use, connect with non-sensitive example data, inspect the enabled tools, and compare responses with their original source. Changes to the service or your client may require a fresh review. The review date is not an uptime guarantee.

An MCP client. The project requires Node.js 24 or newer or Bun 1.4.0 or newer locally. A BLS key is optional; register with BLS for larger quotas.

Configure local stdio with the npm package and MCP_TRANSPORT_TYPE=stdio; optionally set BLS_API_KEY. Remote endpoint: https://bls-labor.caseyjhand.com/mcp. Pin a reviewed version before production rather than relying on @latest.

npx -y @cyanheads/bls-labor-mcp-server@latest

BLS_API_KEY is optional for basic local operation. MCP access and BLS API credentials are separate layers; secure self-hosted HTTP appropriately.

Search BLS series, list surveys and read observations. Optional DataCanvas features inspect staged tables; dropping those tables is disabled by default.

Queries contact BLS directly or pass through the community host. Local operation can create a SQLite catalog cache; optional DataCanvas results are staged. Responses enter the AI context.

Do not present it as an official BLS server. Check series identifiers, seasonal adjustment, units and data vintage. Do not send personal employee data; protect local caches and HTTP access.