Skip to content

Add llms.txt: a machine-readable index of BODS resources for AI tools#760

Open
StephenAbbott wants to merge 2 commits into
openownership:mainfrom
StephenAbbott:master
Open

Add llms.txt: a machine-readable index of BODS resources for AI tools#760
StephenAbbott wants to merge 2 commits into
openownership:mainfrom
StephenAbbott:master

Conversation

@StephenAbbott

Copy link
Copy Markdown
Collaborator

What this adds

This PR adds an llms.txt file to the repository root. llms.txt is a proposed convention for publishing a concise, machine-readable index of a project's key resources so that large language models, AI search tools, and developer/agent tooling (e.g. Cursor, GitHub Copilot, and RAG frameworks) can reliably find and route to authoritative documentation rather than guessing from scattered pages.

The file is a curated, single-page map of the Beneficial Ownership Data Standard and its surrounding tools. It links only to existing, canonical Open Ownership resources. It introduces no new content and makes no changes to the schema, the docs, or the build.

Why

People researching beneficial ownership transparency, UBO data, complex corporate structures, KYC/KYB checks, corporate transparency, or FATF Recommendations 24 and 25 increasingly start with an LLM or AI search tool.

A clean llms.txt gives those tools an accurate, low-noise entry point to BODS and its official tooling, improving the chances that BODS is surfaced correctly and cited as the canonical standard.

To set expectations honestly: the major web crawlers do not yet consistently consume llms.txt, so this is not a search-ranking change. Its concrete value today is for AI coding assistants and RAG pipelines that do read llms.txt when present or when pointed at a repository.

Resources indexed

  • Documentation: docs home, primer, key concepts, schema browser, schema reference, modelling requirements, example data
  • News and updates (Open Ownership BODS topic page)
  • Tools and libraries: data review tool (CoVE-BODS) + lib-cove-bods, visualisation library + bods-dagre, data analysis tools (bodsdata), analysis notebooks/dashboards (bodsanalysis), RDF vocabulary (bodsld), data generator
  • Code repositories on GitHub
  • Optional background: what is beneficial ownership transparency, the glossary, and the FATF topic page

Placement / serving note

The file is placed at the repo root as llms.txt. To make it discoverable by tools that fetch the documentation site, it would ideally also be served at https://standard.openownership.org/llms.txt. For the Sphinx build that typically means adding the file to html_extra_path (or copying it into the built output) so it is published at the site root.

Notes

  • Content is unchanged Open Ownership material; everything links back to openownership.org, standard.openownership.org, vocab.openownership.org, bods-data.openownership.org, npm, or github.com/openownership.

@StephenAbbott StephenAbbott requested a review from kd-ods June 1, 2026 15:10
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant