Skip to main content
For indexer authors and operators, in any language: the guide says which ledger fields stellar-core checks and which it passes through, how to cap recursive structures, how to take participants from the meta, and how to handle muxed and signed-payload addresses. Its checklist turns that into items you can audit your code against. This page says where the guide lives and how it is kept consistent with the corpus and the linter. Outline and review state maps each topic to its section, counts the checklist’s items by mistake, and says how many claims have been reviewed. Verification report says how each claim was checked.
The guide has not yet been reviewed by a stellar-core or platform engineer, and is not published on developers.stellar.org until it has.

How it stays consistent

  • The field-by-field tables are generated from corpus/fields.yaml joined with the corpus manifest, so the table agrees with the corpus by construction; make verify-guide OFFLINE=1 fails when it does not.
  • Every normative sentence carries a claim tag, and make repo-checks fails on a sentence with “must”, “never”, “should” or “do not” that has none.
  • The checklist and the linter share one registry: every xdrlint rule has a checklist item, and the checklist says how to check by hand what the linter cannot see.
  • make verify-guide also runs the stellar-docs repository’s own format and link checks on the page, in a clone at the pinned commit.

Review before publication

Publication waits on a review by a core or platform engineer: every fact claim confirmed or corrected, none disputed. The reviewer package is built by guide/review/build.py. The current state is on Outline and review state.
  • Lint: the same three mistakes as code checks for Go.
  • The judge: participants and balances from the meta, the rule the guide recommends.