Feat/rag knowledge engine - #2
Merged
Merged
Conversation
Introduce a consolidated Streamlit UI in app.py with sample fallback, stateful analysis, and parsing showcase, and add streamlit to requirements for reproducible setup. Co-authored-by: Cursor <cursoragent@cursor.com>
Condense README for faster onboarding with a demo-first structure, and include Dockerfile/.dockerignore to support containerized runs. Co-authored-by: Cursor <cursoragent@cursor.com>
Rename PixelQA references to Agentic Testing Framework across configs, comments, and docs. Document inference backends and normalized outputs in a collapsible DX block. Restructure README with details/summary for scanability. Introduce ATF_MONITOR_MEMORY for tracemalloc profiling; keep PIXELQA_MONITOR_MEMORY as a legacy alias. Update Architecture.md and Streamlit app titles. Add project description in pyproject.toml. Co-authored-by: Cursor <cursoragent@cursor.com>
Add LICENSE (MIT) and CONTRIBUTING.md with local setup, pytest and Ruff commands aligned to CI, optional agent smoke run, and PR expectations. Link new narrative docs from README and document how to contribute and license the project. Co-authored-by: Cursor <cursoragent@cursor.com>
Remove IntegrationGuide, AdvocacyCaseStudy, InterviewNarratives, and linkedin-self-healing-vision-qa from version control while keeping files locally; gitignore those paths. README now links only to Architecture.md for public docs. Add a repository rule so agents do not push narrative or interview markdown unless explicitly requested. Co-authored-by: Cursor <cursoragent@cursor.com>
Run ruff check src tests in GitHub Actions instead of a partial path list. Fix F541/F841/E701 in orchestrator and llama_analyst, and drop an unused import in verify_capture_success. Align CONTRIBUTING and README local CI instructions with the workflow. Co-authored-by: Cursor <cursoragent@cursor.com>
Add golden unit tests for benchmark_evaluator (scores, ranking order, release gate), log_analyzer sliding window, vision_math on synthetic PNGs, and image_validator exposure paths. Declare opencv-python-headless in requirements so CI can import cv2. README Core guarantees now describe test emphasis; roadmap unchecked items become an explicit backlog with rationale for single-thread reproducibility and scoped OpenCV work. Co-authored-by: Cursor <cursoragent@cursor.com>
Ship assets/demo.png (baseline vs AI pipeline UI) and embed it in the Demo preview section; keep optional GIF note for a future recording. Co-authored-by: Cursor <cursoragent@cursor.com>
Update README image embed to match the new filename. Co-authored-by: Cursor <cursoragent@cursor.com>
…tory. Declare full runtime dependencies in pyproject.toml (mirrors requirements.txt), add setuptools src layout (package-dir, py-modules, package __init__ files), and optional dev extras. CI installs with pip install -e ".[dev]"; README and CONTRIBUTING document editable vs requirements.txt. Bump opencv-python-headless to 4.13.0.92 so numpy 2.4.4 resolves cleanly. Rewrite verify_capture_success to write a real temp JPEG and call calculate_metrics with a file path; make existence check deterministic. Dockerfile: default python slim (amd64-friendly), drop incorrect Metal build flags for Linux, and clarify CPU vs macOS Metal in comments. Co-authored-by: Cursor <cursoragent@cursor.com>
Make requirements.txt a one-line -e .[dev] shim; document that only pyproject.toml holds version pins. Docker installs via pip install . from pyproject. CONTRIBUTING adds dependency policy and Streamlit run instructions. app.py: prefer agent.orchestrator with PYTHONPATH=src, fallback to src.agent; add logging for import path, pipeline failures, and bad uploads; widen Ruff to include app.py in CI and docs. Co-authored-by: Cursor <cursoragent@cursor.com>
Add module loggers and INFO/WARNING/ERROR/exception calls in ai_quality_agent, orchestrator, mock_device, vision_math, failure_memory, llama_analyst, log_analyzer, verify_capture_success, test_failure_memory_retrieval, and test_connection. Configure basicConfig in __main__ blocks and at CLI entry (message-only format). Streamlit app.py configures logging when no root handlers exist. Mock device demo code moves under if __name__ guard; widen Ruff to test_connection.py. Remove stale commented print in image_validator. Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
Add util.cli_logging.configure_cli_logging using asctime, levelname, and logger name in the format string; switch batch entrypoints and app.py to use it. test_connection adjusts sys.path in __main__ then imports the helper. Update CONTRIBUTING to describe the new default format. Co-authored-by: Cursor <cursoragent@cursor.com>
Replace deprecated getdata() in vision_math with get_flattened_data for Pillow 14 readiness. Add --cov-fail-under=32 to pytest addopts and document in CONTRIBUTING. Co-authored-by: Cursor <cursoragent@cursor.com>
…aph. Add mypy to optional dev deps and [tool.mypy] (Python 3.11). CI runs two passes: mypy src then MYPYPATH=src on app.py and test_connection.py to avoid duplicate module mapping. Drop the src.agent orchestrator fallback in app.py so only agent.* is used with PYTHONPATH=src (matches docs). Annotate verify_capture_success report dict for mypy. Raise --cov-fail-under to 34. Update CONTRIBUTING for mypy and coverage. Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
Extract httpx-based async backends, ProcessPoolExecutor metrics, and setup docs so batches can overlap HTTP inference without blocking on Pillow CPU work. Co-authored-by: Cursor <cursoragent@cursor.com>
Add startup health checks for LLM planner mode, CLI overrides for planner behavior, typed agent decision trace output with step-level fallback visibility, and contract-aligned tests/docs so report artifacts distinguish true LLM planning from simulated fallback. Co-authored-by: Cursor <cursoragent@cursor.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.