ERD_INDEXERS_CHECKLIST.md still had a reopened CLI parity gap after the read/list slice landed.
Operators could inspect indexer resources from the CLI, but tag lifecycle, secret lifecycle, and category-mapping writes still required the UI or raw API calls.
The next efficient step needed to reuse the existing stored-proc-backed HTTP surface instead of adding new runtime behavior.
Decision:
Extend revaer-cli with indexer tag, indexer secret, and indexer category-mapping subcommands that call the existing /v1/indexers/... endpoints.
Keep the scope focused on operator write parity for tags, secrets, tracker category mappings, and media-domain mappings, with targeted CLI integration tests that assert exact request paths and payloads.
Leave the broader CLI parity checklist item open until routing-policy, rate-limit, search-profile, backup/restore, and RSS mutation flows also exist.
Consequences:
Operators can now manage common indexer metadata and mapping writes from the CLI without dropping to raw HTTP.
The implementation stays dependency-light by reusing the existing reqwest client and output layer.
CLI parity is still incomplete overall, so the checklist must continue to call out the remaining mutation surfaces explicitly.
Follow-up:
Add CLI write coverage for routing policies, rate limits, search profiles, and backup/restore.
Add CLI mutation flows for RSS state and any remaining category/profile assignment surfaces needed for full ERD parity.