Documentation index

This directory and the repo root hold all user- and developer-facing documentation. Start with the top-level README for the product overview and install path.

Start here

• ../README.md — what the project is, install, run.
• ../CONTRIBUTING.md — branching, tests, PR expectations.
• ../SECURITY.md — how to report vulnerabilities.
• ../CHANGELOG.md — release history.

Architecture

• ../ARCHITECTURE.md — layers, responsibilities, patterns, data flow.
• ../ARCHITECTURE_DIAGRAM.md — visual representation of the module structure.
• superpowers/specs/ — historical design specs (e.g. recurring-bills classification).

Operations

• ../DEPLOYMENT_GUIDE.md — step-by-step Streamlit Cloud deployment.
• ../DEPLOYMENT_OPTIONS.md — comparison of hosting options.
• deploy/ — VPS systemd blueprints (knesset-etl.service, knesset-etl.timer).

Research

• CAP_ANNOTATION_GUIDE.md — CAP bill-classification annotation workflow.
• classification_process.md — recurring-bills classification method and the sendable Prof. Amnon workbook contract.
• QUERY_EXPORTS.md — generating and validating the sendable all-parliamentary-queries CSV.
• FULL_DATASET_DOWNLOAD.md — exporting a full filtered query result.
• ../PERFORMANCE_OPTIMIZATIONS.md — caching strategies and bottleneck notes.
• DATA_PROVENANCE.md — where the data comes from, how it is refreshed, licensing notes.

Reference

• KnessetOdataManual.pdf — the upstream Knesset OData API reference, for schema-level questions about source tables.

Developer memory

• ../CLAUDE.md — non-obvious rules the AI assistant and maintainers rely on (DuckDB sequence patterns, faction-matching SQL, the Parquet snapshot contract, Streamlit PYTHONPATH quirk). Gitignored by intent but present in every clone as authoritative internal guidance.