v0.3.1

Highlights

• Added multimodal ScreenPipe context ingestion, timeline thumbnails, and event-driven background observation curation.
• Expanded the Data Audit experience with guided speaker enrollment, background review/suppression tooling, richer transcript editing, and the Chronicle Espresso interface.
• Unified the desktop companion controls for ScreenPipe, vault sync, and pendant capture; simplified client configuration around the repository-root .env; and fixed pendant Opus loading.
• Improved the mobile app's connection experience with live backend reachability on the home screen, pull-to-refresh, and clearer offline/backend-down diagnosis.
• Expanded backend test coverage and CI diagnostics, restored speaker-recognition integration CI, and fixed several session, worker-startup, timestamp, and runtime-configuration issues.
• Added a canonical release runbook covering validation, TestFlight, GitHub publication, container images, and post-release verification.

Operational note

Desktop client configuration is now consolidated in the repository-root .env. Existing desktop companion users should review the updated setup guidance when upgrading.

Validation

• Candidate: dev at de09293bd1e5f92ebe11d7d34a4dc9492409e523
• Python tests: passed (workflow run)
• Full Robot integration suite: 154 passed, 13 failed (workflow run). The remaining failures cover memory/search integration, legacy Redis session-schema expectations, one transcript-timestamp assertion, and streaming completion.
• Speaker recognition integration: passed (workflow run)
• iPhone/TestFlight: signed EAS build 1.0.8 (41) completed and was accepted by Apple App Store Connect (workflow run, submission details). Apple processing is in progress.

What's Changed

• Repository cleanup, workflow coverage, and screenshot tooling by @AnkushMalaker in #336
• feat(scripts): display available disk space in cleanup stats table by @omer-said-bayhan in #337
• Guide ScreenPipe capture-node setup in wizard by @AnkushMalaker in #339
• check by @omer-said-bayhan in #338
• Feat/screenpipe wizard by @AnkushMalaker in #340
• chore: share agent skills across tools by @AnkushMalaker in #341
• fix(speaker-recognition): restore green integration CI by @AnkushMalaker in #343
• fix(tray): unify wearable client on repo-root .env + fix pendant Opus load by @AnkushMalaker in #344

New Contributors

• @omer-said-bayhan made their first contribution in #337

Full Changelog: v0.3.0...v0.3.1

Docker Images

All images are published to ghcr.io/simpleopensoftware with tags v0.3.1 and latest.

Core Services

docker pull ghcr.io/simpleopensoftware/chronicle-backend:v0.3.1

Parakeet ASR (pick your CUDA version)

docker pull ghcr.io/simpleopensoftware/chronicle-asr-nemo-cu126:v0.3.1
docker pull ghcr.io/simpleopensoftware/chronicle-asr-nemo-cu128:v0.3.1

Speaker Recognition (pick your variant)

docker pull ghcr.io/simpleopensoftware/chronicle-speaker-cpu:v0.3.1
docker pull ghcr.io/simpleopensoftware/chronicle-speaker-cu126:v0.3.1
docker pull ghcr.io/simpleopensoftware/chronicle-speaker-cu128:v0.3.1