Last reviewed: 2026-07-12. Base branch: main.
This file is a plain-language map of the project as it exists on main. It does not replace the README, roadmap, audit docs, or source comments. It points to them so a reviewer can see the whole shape without reading every file first.
Olive's Bark Logger is an on-device noise monitor and report generator. It measures sound levels and event timing, stores only derived event data, and produces reports without recording audio.
Package metadata checked in this pass:
- Python package
olive-bark-loggerfor Python>=3.9.
- People who need a factual log of barking or noise patterns in their home.
- Maintainers building privacy-preserving measurement tools.
- Reviewers checking that no audio is stored, transmitted, or implied by the reports.
- Level calculation, event detection, SQLite storage, report rendering, and quiet-hours summaries.
- A Raspberry Pi service path and a browser PWA variant.
- Calibration and tuning CLIs.
- Docs for methodology, gaps, audits, ADRs, PWA use, and deployment.
- Tests for no-audio behavior, no egress, reporting, detector logic, and PWA code.
- monitor/ contains capture, level, detector, health, config, and service code.
- report/ contains charts, aggregation, export, and rendering.
- store/ contains the SQLite layer.
- pwa/ holds the browser variant.
- docs/ records audits, gaps, ADRs, methodology, and limitations.
Observed source and operations surfaces:
DockerfileMakefiledeploy/monitor/pwa/pyproject.tomlreport/scripts/store/
GitHub workflow files checked:
.github/workflows/ci.yml
- The central guarantee is no audio bytes written to disk or sent out.
- Reports state limits: uncalibrated dBFS is relative, and the tool cannot prove what made a sound.
- Local-only operation reduces disclosure risk, but users still control where exported reports go.
- It is not a surveillance device.
- It cannot identify a sound source with certainty.
- It does not make legal or lease-compliance conclusions.
This pass checked 37 hand-authored doc or metadata files, 33 Python/Node test files, and 1 workflow file on main. The count excludes vendored provider licenses, dependency folders, generated cache files, and generated artifacts.
Primary docs checked:
.github/PULL_REQUEST_TEMPLATE.md.github/rulesets/README.mdCHANGELOG.mdCITATION.cffCONTRIBUTING.mdDEFINITION_OF_DONE.mdLICENSEREADME.mdSECURITY.mddocs/GAP-LEDGER.mddocs/I18N.mddocs/RESPONSIBLE-TECH-AUDITS.mddocs/RESEARCH-ROADMAP.mddocs/ROADMAP.mddocs/USER-RESEARCH.mddocs/ideation/README.mddocs/a11y/STATEMENT.mddocs/adr/0000-record-architecture-decisions.mddocs/adr/0001-single-maintainer-review-posture.mddocs/adr/0002-python-39-floor.mddocs/audits/accessibility-2026-06-05.mddocs/audits/data-card.mddocs/audits/methodology-and-limitations.mddocs/audits/no-audio-guarantee.mddocs/audits/recording-law-notes.mddocs/audits/residual-risk.mdpwa/README.mdspec/SEMANTICS.md
Representative test files checked:
pwa/detector.test.mjspwa/report.test.mjstests/conftest.pytests/snapshots/report.htmltests/test_a11y.pytests/test_aggregate.pytests/test_calibrate.pytests/test_capture_resilient.pytests/test_cli.pytests/test_config.pytests/test_detector.pytests/test_eval.pytests/test_export.pytests/test_features.pytests/test_health.pytests/test_heatmap.pytests/test_integration.pytests/test_level.pytests/test_no_audio.pytests/test_no_egress.pytests/test_properties.pytests/test_report_content.pytests/test_report_snapshot.pytests/test_store_durability.pytests/test_violations.py
For this docs PR, validation means the scope file was generated from the clean origin/main worktree, reviewed against repo metadata and docs inventory, and checked with git diff --check. Project test suites are still the authority for code behavior, because this PR changes documentation only.