Releases: sestinj/threader
Releases · sestinj/threader
v0.0.25
v0.0.24
v0.0.23
What's Changed
Full Changelog: v0.0.22...v0.0.23
v0.0.22
What's Changed
Full Changelog: v0.0.21...v0.0.22
v0.0.21
Full Changelog: v0.0.20...v0.0.21
v0.0.20
Full Changelog: v0.0.19...v0.0.20
v0.0.19
Full Changelog: v0.0.18...v0.0.19
v0.0.18
What's New
threader debug subcommand
New diagnostic tool for troubleshooting transcript sync issues between local Claude Code sessions and the cloud.
Subcommands:
threader debug list— Overview of all sessions with source/local/synced line counts, queue depth, and agethreader debug inspect <session_id>— Detailed metadata, transcript paths, sync state, and pending queue entries (supports prefix matching)threader debug verify <session_id>— Compare transcripts across source, local copy, and remote with SHA-256 hashing, duplicate detection, and image-transform awarenessthreader debug diff <session_id>— Line-by-line diff between local and remote with color-coded output
Full Changelog: v0.0.17...v0.0.18
v0.0.17
Bug Fix
- Fix all threads showing 0 tokens: Token counting was reading from
~/.claude/__store.dbwhich uses different session IDs than the Claude Code hooks. Switched to parsing usage data directly from the source transcript JSONL files. - Live token updates during active sessions: Added
refresh_cost()to the transcript poller so token counts update every 5 seconds during active sessions, not just on Stop/End events. - Removed
rusqlitedependency: No longer needed, reducing binary size and compile time.
Full Changelog: v0.0.16...v0.0.17
v0.0.16
What's New
- Faster mid-turn sync: Added a transcript poller that checks active sessions every 5 seconds, so conversation content syncs to the cloud during agent turns rather than only after they end.
Full Changelog: v0.0.15...v0.0.16