{"id":50654662,"url":"https://github.com/tiroq/k6-visual-inspector","last_synced_at":"2026-06-07T23:03:34.918Z","repository":{"id":358012949,"uuid":"1239484745","full_name":"tiroq/k6-visual-inspector","owner":"tiroq","description":"A small toolkit for post-processing `k6/browser` screenshots after load or UI test runs.","archived":false,"fork":false,"pushed_at":"2026-05-15T07:39:47.000Z","size":3791,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-05-15T09:37:59.403Z","etag":null,"topics":["failure-analysis","k6","k6-browser","load-testing","ocr","opencv","python","qa-automation","screenshot-analysis","screenshot-clustering","tesseract","test-automation","ui-testing","visual-regression","visual-testing"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/tiroq.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-05-15T06:19:35.000Z","updated_at":"2026-05-15T07:39:51.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/tiroq/k6-visual-inspector","commit_stats":null,"previous_names":["tiroq/k6-visual-inspector"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/tiroq/k6-visual-inspector","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tiroq%2Fk6-visual-inspector","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tiroq%2Fk6-visual-inspector/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tiroq%2Fk6-visual-inspector/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tiroq%2Fk6-visual-inspector/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/tiroq","download_url":"https://codeload.github.com/tiroq/k6-visual-inspector/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tiroq%2Fk6-visual-inspector/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34041091,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-07T02:00:07.652Z","response_time":124,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["failure-analysis","k6","k6-browser","load-testing","ocr","opencv","python","qa-automation","screenshot-analysis","screenshot-clustering","tesseract","test-automation","ui-testing","visual-regression","visual-testing"],"created_at":"2026-06-07T23:03:34.255Z","updated_at":"2026-06-07T23:03:34.908Z","avatar_url":"https://github.com/tiroq.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# k6 Visual Inspector\n\nA small toolkit for post-processing `k6/browser` screenshots after load or UI test runs.\n\n![icon](./assets/image.png)\n\nIt groups visually similar screenshots, extracts OCR text, detects basic UI composition, and generates a compact report so you can quickly understand repeated UI failures instead of reviewing hundreds of nearly identical images manually.\n\n## What it does\n\n- Clusters screenshots by visual similarity, OCR text, and UI layout composition.\n- Detects common UI patterns such as dialogs, content frames, headers, empty states, loading screens, and browser errors.\n- Extracts text from full screenshots, central areas, and detected regions.\n- Produces representative screenshots for each cluster.\n- Generates HTML, JSON, JSONL, CSV, and cluster folders for manual triage.\n- Supports parallel processing for large screenshot sets.\n\n## Use case\n\nThis project is useful when a `k6/browser` run produces many screenshots, and most of them look similar or differ only by dynamic text such as IDs, timestamps, counters, or request-specific messages.\n\nInstead of reviewing screenshots one by one, you get a grouped report like:\n\n```text\ncluster-000 error_modal.backend_500 count=143 severity=high\ncluster-001 loading.timeout count=87 severity=high\ncluster-002 empty_state.no_data count=41 severity=low\ncluster-003 browser_error.bad_gateway_502 count=12 severity=high\n```\n\n## Install\n\n```bash\npython3 -m venv .venv\nsource .venv/bin/activate\n\npip install -r requirements.txt\n```\n\nTesseract is required for OCR:\n\n```bash\n# macOS\nbrew install tesseract\n\n# Debian / Ubuntu\nsudo apt-get install -y tesseract-ocr tesseract-ocr-eng\n```\n\nOptional Russian OCR:\n\n```bash\nsudo apt-get install -y tesseract-ocr-rus\n```\n\n## Usage\n\n```bash\npython3 cluster_screenshots.py ./screenshots ./analysis\n```\n\nRecommended first run:\n\n```bash\npython3 cluster_screenshots.py ./screenshots ./analysis \\\n  --ocr-engine tesseract \\\n  --ocr-mode fast \\\n  --workers auto\n```\n\nBalanced OCR mode:\n\n```bash\npython3 cluster_screenshots.py ./screenshots ./analysis-balanced \\\n  --ocr-engine tesseract \\\n  --ocr-mode balanced \\\n  --workers auto\n```\n\nDebug OCR crops:\n\n```bash\npython3 cluster_screenshots.py ./screenshots ./analysis-debug \\\n  --ocr-mode fast \\\n  --debug-ocr\n```\n\n## Outputs\n\n```text\nanalysis/\n  report.html\n  clusters.json\n  items.jsonl\n  cluster-summary.csv\n  similarity_visual.npy\n  similarity_text.npy\n  similarity_layout.npy\n  similarity_rule.npy\n  similarity_combined.npy\n  overlays/\n  clusters/\n  debug-ocr/\n```\n\nThe main file to open is:\n\n```text\nanalysis/report.html\n```\n\n## Status\n\nExperimental but practical. The current approach is intentionally local-first and explainable: OpenCV, image hashes, OCR, rule-based labels, and clustering. LLM or embedding-based labeling can be added later as an optional layer on top of representative clusters.","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftiroq%2Fk6-visual-inspector","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ftiroq%2Fk6-visual-inspector","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftiroq%2Fk6-visual-inspector/lists"}