What Dc Code does
The Dc Code MCP server exposes current District of Columbia statutes through two purpose-built tools. It is intended for the ⚖️ Legal category and supports citation lookup, topic discovery, and browsing sections within a title.
dc_statute retrieves one section using a citation such as 22-2101. Its response can include the official heading, current statutory text, disposition, legislative history, prior codifications, and cross references. Dispositions identify sections that are in force, repealed, transferred, or reserved.
dc_search searches the official headings of D.C. Code sections. Queries can target subjects such as eviction notices or concealed-carry permits, and an optional title limits the results. Supplying a title without a query lists the sections in that title, providing a chapter or title browsing mode.
How it works
The Dc Code MCP server uses the District of Columbia Council’s law-xml-codified repository as its source. Section text is fetched live from the current publication branch selected during the index-building process. The search index contains section headings and structural information rather than the statute body, so search results are not full-text matches.
The baked index covers the available section headings and is refreshed manually when the publisher releases new amendment branches. The live statute request remains separate from that index, which means a section lookup can retrieve current text even when the local heading index needs rebuilding.
History annotations may identify enacting or amending Acts, but the source does not expose a separate canonical effective-date field. Historical versions before amendments are not available because the upstream repository publishes current codified text only.
Setup and configuration
The Dc Code MCP server can be connected through the Pipeworx gateway with this endpoint:
{
"mcpServers": {
"dc-code": {
"url": "https://gateway.pipeworx.io/dc-code/mcp"
}
}
}
It also runs locally as a stdio package without a Pipeworx account:
{
"mcpServers": {
"dc-code": {
"command": "npx",
"args": ["-y", "@pipeworx/mcp-dc-code"]
}
}
}
The gateway endpoint includes shared Pipeworx tools in addition to the pack-specific tools. The standalone package exposes only the D.C. Code tools. Both modes are keyless. The gateway also offers direct HTTP tool calls, including POST /v1/tools/dc_statute.
Tools and capabilities
dc_statute(citation): retrieves the current XML-derived content and grouped annotations for one D.C. Code section.
dc_search(query?, title?, limit?): searches official headings, optionally restricts results to a title, and lists a title’s sections when no query is supplied.
- Citation lookup does not require an index lookup because the citation identifies the title used to construct the source URL.
- Gateway users can use
ask_pipeworx to submit a plain-language question that routes to an appropriate tool.
Limitations and notes
Search covers official headings only; it is not a search across the full statutory text. The index can become stale as the Council publishes amendment branches, and updating it requires manually running the repository’s bake script. dc_statute fetches text live using the branch resolved by the most recent bake.
This is an independent, unofficial integration. It uses the Council’s public-domain code content through the Council’s GitHub publication channel and does not retrieve content from code.dccouncil.gov directly.