main

[docs] split the json frame mapping out of the chunk event format

refetch-budget-v1

[docs] describe refetch budgets and rewalking newer repos

__dolt_remote_info__

dolt remote info

firehose-restart-jitter

[firehose] test that requestCrawl restarts a source only after it gave up

backup/main-knotrs-20260922

[backfill] name the unread transition states

`BackfillImported` carries the before and after states so `emit` can compare them, but `emit` only exists with the stream feature. The indexer-only and indexer-backlinks-nostream builds therefore warned on both fields, which failed the feature matrix for everyone.

identity-recovery

[ssrf] couple public egress validation to sockets

record-bodies

[db] write v10 bodies into the compressed heads keyspace

v9 created records with compression disabled at every level and fjall persists keyspace options on disk, so bodies written there stay uncompressed no matter what the schema says at open (measured on the prod db copy: 71.3MB stored vs 16MB zstd:3 standalone on the same bodies). the schema registers the store under a new name, heads, with zstd at every level, and v10 stages rewritten bodies straight into it, retiring both legacy stores in its finalize. chunked passes now resolve scan keyspaces through an on-disk fallback for stores the registry has dropped, pass the scanned keyspace to visit so in-place rewrites keep writing the store they read, and treat a never-existing source as a no-op when the pass is optional_on_absent. verified on a copy of the prod database: 136,311 entries migrated in ~500ms, heads compacts to 13.1MB (5.4x smaller), record resolution unchanged.

opt-work

awa

hydrant-9ht

[backfill] stream getRepo with a hard body bound

hydrant-8f0

[backfill] batch full CAR block access

fix-counts

[db] refactor how repo counts are handled

test-13-fix

wip