Skip to content
This repository was archived by the owner on May 13, 2026. It is now read-only.
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
37 changes: 0 additions & 37 deletions .github/workflows/notify-downstream.yml

This file was deleted.

4 changes: 2 additions & 2 deletions CLAUDE.md
Original file line number Diff line number Diff line change
Expand Up @@ -31,9 +31,9 @@ npm run generate:api # Generate TypeScript types from OpenAP

## Bump cascade (downstream)

When a PR merges to `main`, `.github/workflows/notify-downstream.yml` fires a `repository_dispatch` to `EdgeVector/schema_service`, which auto-bumps its `fold_db` rev pin and merges. That cascades further into `EdgeVector/fold_db_node`. End-to-end ~10–15 min from a merge here to fold_db_node landing.
Downstream consumers pull this repo's main HEAD on a **2-hour schedule**: `EdgeVector/schema_service`'s `bump-fold-db.yml` (cron) bumps its fold_db rev pin, then `EdgeVector/fold_db_node`'s `bump-schema-service.yml` (cron) picks up schema_service's tip plus the matching fold_db rev. End-to-end cascade lag: up to 2h schema_service hop + up to 2h fold_db_node hop. The cascade used to be dispatch-driven from this repo via `notify-downstream.yml`, but per-merge bumps superseded each other in downstream merge queues — switched to scheduled polling 2026-05-01.

If you're shipping a breaking fold_db change that requires consumer Rust updates alongside the rev bump, disable the relevant `bump-*.yml` workflow in `EdgeVector/schema_service` and `EdgeVector/fold_db_node` Actions before merging here, do the manual consumer PRs, re-enable.
If you're shipping a breaking fold_db change that requires consumer Rust updates alongside the rev bump, disable the relevant `bump-*.yml` workflow in `EdgeVector/schema_service` and `EdgeVector/fold_db_node` Actions before merging here, do the manual consumer PRs, re-enable. Force-flush an immediate bump cycle (rather than waiting up to 2h) via `gh workflow run bump-fold-db.yml -R EdgeVector/schema_service` and `gh workflow run bump-schema-service.yml -R EdgeVector/fold_db_node` after this lands.

## AI Provider Configuration

Expand Down
Loading