dbook 0.1.0__tar.gz

dbook-0.1.0/.github/workflows/ci.yml

@@ -0,0 +1,55 @@
+name: CI
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+    branches: [main]
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version: ["3.11", "3.12"]
+
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Set up Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v5
+        with:
+          python-version: ${{ matrix.python-version }}
+
+      - name: Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install -e ".[dev]"
+
+      - name: Run tests
+        run: |
+          pytest tests/ -q --tb=short --no-header
+
+      - name: Build package
+        run: |
+          pip install build
+          python -m build --sdist
+
+  lint:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+
+      - name: Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install ruff
+
+      - name: Ruff check
+        run: |
+          ruff check src/ --output-format concise --quiet

dbook-0.1.0/.github/workflows/publish.yml

@@ -0,0 +1,28 @@
+name: Publish to PyPI
+
+on:
+  release:
+    types: [published]
+
+jobs:
+  publish:
+    runs-on: ubuntu-latest
+    environment: pypi
+    permissions:
+      id-token: write  # Required for trusted publishing
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+
+      - name: Install build tools
+        run: pip install build
+
+      - name: Build package
+        run: python -m build
+
+      - name: Publish to PyPI
+        uses: pypa/gh-action-pypi-publish@release/v1

dbook-0.1.0/.gitignore

@@ -0,0 +1,12 @@
+__pycache__/
+*.egg-info/
+*.pyc
+.eggs/
+dist/
+build/
+.pytest_cache/
+.ruff_cache/
+*.egg
+.claude/
+uv.lock
+REPORT.html

dbook-0.1.0/CLAUDE.md

@@ -0,0 +1,105 @@
+# dbook — Development Rules
+
+## What This Project Is
+A metadata compiler that introspects databases and generates a layered directory of markdown files for AI agent consumption. The output format IS the product — agents read these files to understand databases.
+
+## Golden Rules
+
+### 1. Never sacrifice output quality to pass a test
+If a test has unrealistic targets, fix the test — do NOT strip content from the output to make the number look better. The output must be genuinely useful for an AI agent. A compact but useless output is worse than a verbose but informative one.
+
+### 2. The output format is a UX decision
+Every markdown file an agent reads is a user interface. Treat it with the same care as a frontend component:
+- NAVIGATION.md MUST use markdown tables for schema listings (not bullet lists)
+- Table .md files MUST show all columns with Type, Nullable, Default, PK, and Comment
+- Sample data MUST show 3-5 rows with ALL columns (not truncated)
+- Foreign keys MUST show full qualified references
+- Referenced By sections MUST use schema-qualified names
+
+### 3. Benchmark against realistic scale
+- The primary test fixture (13 tables) tests correctness
+- Token savings benchmarks MUST also run against a scaled fixture (50+ tables)
+- All 10 benchmark questions must be tested, not a subset
+- Per-question token breakdown must be reported
+- Baseline = tokens for ALL raw DDL; per-question cost = tokens agent actually reads
+
+### 4. Progressive disclosure is the value proposition
+Token savings come from the ARCHITECTURE (read 2-3 files instead of all tables), not from compacting individual files. If NAVIGATION.md is 150 tokens and a table file is 200 tokens, that's fine — the savings come from NOT reading the other 49 table files.
+
+### 5. Don't over-filter the concept index
+The concept index should include ALL meaningful terms extracted from table and column names. Filtering out "common" terms like "id", "name", "type" removes the most-searched terms. Only filter truly structural noise: single characters, articles.
+
+### 6. Test what matters
+- E2E tests: verify output structure and content completeness
+- Benchmark tests: measure token efficiency at realistic scale AND verify agent can find correct answers
+- No unit tests in isolation — always test the full pipeline
+
+## Output Format Specifications
+
+### NAVIGATION.md (L0 — must be <300 tokens)
+```
+# Database Book: {dialect}
+Compiled: {timestamp} | Mode: {mode}
+
+## Schemas
+
+| Schema | Tables | Total Rows | Description |
+|--------|--------|-----------|-------------|
+| auth | 4 | 60,205 | users, sessions, roles, user_roles |
+| billing | 6 | 138,550 | orders, invoices, payments, products, ... |
+
+## Quick Reference
+- User data: schemas/auth/users.md
+- Financial: schemas/billing/_manifest.md
+
+## How to Navigate
+1. Read this file for overview
+2. Check concepts.json to find specific terms
+3. Read schemas/{name}/_manifest.md for schema details
+4. Read schemas/{name}/{table}.md for full table metadata
+```
+
+### Per-Table .md (L2 — no size limit, completeness matters)
+Must include ALL of these sections when applicable:
+- Header with mechanical summary
+- Columns table (ALL columns, ALL attributes)
+- Primary Key
+- Foreign Keys with full references
+- Indexes
+- Sample Data (3-5 rows, all columns)
+- Referenced By (schema-qualified)
+
+### concepts.json (Ls)
+- Include all terms from splitting table/column names on underscores and camelCase
+- Only filter: single characters, pure numbers
+- Each term maps to tables (file paths) and columns (qualified names)
+- aliases field present but empty in base mode (populated by LLM in Phase 5)
+
+## Architecture
+
+```
+SQLAlchemyCatalog → BookMeta → Compiler → Markdown Directory
+                                  ↓
+                    generators/navigation.py  → NAVIGATION.md
+                    generators/manifest.py    → _manifest.md
+                    generators/table.py       → {table}.md
+                    generators/concepts.py    → concepts.json
+                    generators/checksums.py   → checksums.json
+```
+
+## Commands
+```bash
+pip install -e ".[dev]"          # Install in dev mode
+pytest tests/ -q --tb=short      # Run all tests
+pytest tests/ -v -s              # Run with verbose output (see benchmark reports)
+```
+
+## Key Files
+- src/dbook/models.py — Data models (ColumnInfo, TableMeta, BookMeta, etc.)
+- src/dbook/catalog.py — Catalog protocol + SQLAlchemyCatalog
+- src/dbook/hasher.py — SHA256 schema hashing
+- src/dbook/compiler.py — Compile BookMeta → markdown directory
+- src/dbook/generators/ — Individual markdown/json generators
+- tests/conftest.py — SQLite test fixture (13 tables)
+- tests/benchmark_helpers.py — AgentSimulator, token counting, BenchmarkReport
+- PLAN.md — Full implementation plan (7 phases)

dbook-0.1.0/LICENSE

@@ -0,0 +1,200 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to the Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by the Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding any notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. Please also get an
+      "Alarm" or "alarm.txt" file (see above).
+
+      Copyright [yyyy] [name of copyright owner]
+
+      Licensed under the Apache License, Version 2.0 (the "License");
+      you may not use this file except in compliance with the License.
+      You may obtain a copy of the License at
+
+          http://www.apache.org/licenses/LICENSE-2.0
+
+      Unless required by applicable law or agreed to in writing, software
+      distributed under the License is distributed on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied. See the License for the specific language governing
+      permissions and limitations under the License.

dbook-0.1.0/MANIFEST.in

@@ -0,0 +1,4 @@
+include LICENSE
+include README.md
+include PLAN.md
+recursive-include docs *.svg *.html

dbook-0.1.0/PKG-INFO

@@ -0,0 +1,260 @@
+Metadata-Version: 2.4
+Name: dbook
+Version: 0.1.0
+Summary: Database metadata compiler for AI agent consumption
+Project-URL: Homepage, https://github.com/ShurikM/dbook
+Project-URL: Repository, https://github.com/ShurikM/dbook
+Project-URL: Issues, https://github.com/ShurikM/dbook/issues
+Author-email: Shurik M <shurik@example.com>
+License-Expression: Apache-2.0
+License-File: LICENSE
+Keywords: agent,ai,database,introspection,metadata,schema,sql
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Database
+Classifier: Topic :: Software Development :: Libraries
+Requires-Python: >=3.11
+Requires-Dist: click>=8.0
+Requires-Dist: sqlalchemy>=2.0
+Requires-Dist: sqlglot>=20.0
+Provides-Extra: all
+Requires-Dist: anthropic; extra == 'all'
+Requires-Dist: openai; extra == 'all'
+Requires-Dist: presidio-analyzer; extra == 'all'
+Requires-Dist: presidio-anonymizer; extra == 'all'
+Requires-Dist: pyyaml; extra == 'all'
+Provides-Extra: bigquery
+Requires-Dist: sqlalchemy-bigquery; extra == 'bigquery'
+Provides-Extra: dev
+Requires-Dist: pytest>=7.0; extra == 'dev'
+Requires-Dist: tiktoken; extra == 'dev'
+Provides-Extra: embeddings
+Requires-Dist: sentence-transformers; extra == 'embeddings'
+Provides-Extra: llm
+Requires-Dist: anthropic; extra == 'llm'
+Requires-Dist: openai; extra == 'llm'
+Provides-Extra: metrics
+Requires-Dist: pyyaml; extra == 'metrics'
+Provides-Extra: mysql
+Requires-Dist: pymysql; extra == 'mysql'
+Provides-Extra: pii
+Requires-Dist: presidio-analyzer; extra == 'pii'
+Requires-Dist: presidio-anonymizer; extra == 'pii'
+Provides-Extra: postgres
+Requires-Dist: psycopg2-binary; extra == 'postgres'
+Provides-Extra: snowflake
+Requires-Dist: snowflake-sqlalchemy; extra == 'snowflake'
+Description-Content-Type: text/markdown
+
+# dbook
+
+A database metadata compiler that makes AI agents understand your database — not just its structure, but its meaning.
+
+> **dbook** compiles your database schema into AI-ready metadata — enum values, semantic relationships, example queries, auto-detected metrics, data lineage, and PII markers. In SQL execution benchmarks, agents with dbook produce 100% correct SQL vs 75% with raw DDL.
+
+## The Problem
+
+Your AI agents are **blind to your data**.
+
+Raw DDL tells agents the structure — but not the meaning:
+- `status VARCHAR(20)` — agents guess "active", "enabled", "1"... the real values are "pending", "shipped", "delivered"
+- `user_id INTEGER REFERENCES users(id)` — but what IS this relationship? The customer? The assignee? The creator?
+- Your gold layer exists because consumers couldn't read silver — but AI agents CAN, with the right metadata
+
+**The result:**
+- You maintain expensive gold layer ETL just for AI consumption
+- Every agent re-discovers the schema independently (10 agents = 10x cost)
+- Schema changes break agents silently — no one knows until production fails
+- Agents access PII columns unknowingly — compliance risk with every query
+- Agents guess enum values and write wrong SQL — silent data quality issues
+
+## What dbook Does
+
+Connects to any database, introspects the schema, and generates structured metadata that gives agents the context DDL lacks:
+
+<p align="center">
+  <img src="docs/architecture.svg" alt="dbook Architecture" width="800">
+</p>
+
+```bash
+pip install dbook
+dbook compile "postgresql://user:pass@host/db" --output ./my_dbook
+```
+
+### What agents get:
+
+**1. Enum value documentation** — auto-detected via `SELECT DISTINCT`
+```
+status: pending, confirmed, shipped, delivered, cancelled
+method: credit_card, debit_card, paypal, bank_transfer
+```
+
+**2. Semantic FK descriptions** — agents understand relationships
+```
+→ users via user_id — the customer who placed this order
+← order_items.order_id — line items in this order
+```
+
+**3. Example queries** — patterns agents can follow
+```sql
+- By status: SELECT * FROM orders WHERE status IN ('pending', 'confirmed')
+- Revenue over time: SELECT DATE(created_at), SUM(total) FROM orders GROUP BY DATE(created_at)
+```
+
+**4. Auto-detected metrics** — common aggregations ready to use
+```
+- Total Amount: SELECT SUM(total) FROM orders
+- Count by Status: SELECT status, COUNT(*) FROM orders GROUP BY status
+- Amount over time: SELECT DATE(created_at), SUM(total) FROM orders GROUP BY DATE(created_at)
+```
+
+**5. Data lineage** — how tables connect in the data flow
+```
+Source tables: users, products (no dependencies)
+Intermediate: orders → depends on users | ← used by order_items, invoices
+Leaf: payments → depends on invoices
+```
+
+**6. PII detection** — marks sensitive columns, redacts sample data
+```
+| email | VARCHAR(255) | EMAIL (0.90) | high |
+| card_last_four | VARCHAR(4) | CREDIT_CARD_PARTIAL (0.70) | low |
+```
+
+**7. Query validation** — SQLGlot-powered, catches errors before execution
+```python
+validator = QueryValidator(book)
+result = validator.validate("SELECT * FROM orders WHERE status = 'completed'")
+# Warning: 'completed' not in known values: pending, confirmed, shipped, delivered, cancelled
+```
+
+## Key Benchmark Results
+
+### SQL Execution Benchmark: DDL vs dbook
+
+Tested on an Amazon-like e-commerce database (34 tables, 15 business tasks, 4 agent types):
+
+| Fact Type | Raw DDL | Base dbook | LLM dbook |
+|-----------|---------|-----------|-----------|
+| Structural (column names) | 100% | 100% | 100% |
+| Value-level (enum values) | 21% | 88% | 94% |
+| **Overall key fact coverage** | **76%** | **96%** | **98%** |
+| **SQL execution correctness** | **75%** | **100%** | **100%** |
+
+**In the SQL execution benchmark, dbook achieves 100% correct SQL vs 75% with raw DDL** — the difference between agents that guess enum values and agents that know them.
+
+### On a 5-table database:
+- DDL key fact coverage: 69% -> dbook: 93% (+24% improvement)
+- SQL execution benchmark: DDL produces 75% correct SQL -> dbook: 100% correct SQL
+
+### Agent Discovery (business-term search):
+- 15 real business tasks (billing, sales, support, analytics agents)
+- All 3 modes achieve 15/15 success with mechanical aliases
+- Business terms like "shopping cart", "refund", "A/B test" correctly map to tables
+
+### Token Savings (at scale):
+- 50 tables: ~50% fewer tokens per query vs reading all DDL
+- Scales linearly — larger databases see larger savings
+
+## Architecture
+
+```
+SQLAlchemy Inspector → BookMeta → Compiler → Output Directory
+                                     ↓
+                      NAVIGATION.md    (table overview + lineage)
+                      schemas/
+                        {schema}/
+                          _manifest.md  (schema details + relationships)
+                          {table}.md    (columns, values, FKs, metrics, examples)
+```
+
+### Catalog Protocol
+Database-agnostic via `Catalog` protocol. Default `SQLAlchemyCatalog` supports any SQLAlchemy-compatible database. DB type auto-detected from URL.
+
+### Supported Databases
+PostgreSQL, MySQL, SQLite, Snowflake, BigQuery — any database with a SQLAlchemy dialect.
+
+## Usage
+
+### Full compile
+```bash
+dbook compile "postgresql://user:pass@host/db" --output ./my_dbook
+```
+
+### With PII detection (marks sensitive columns, redacts sample data)
+```bash
+pip install dbook[pii]
+dbook compile "postgresql://..." --output ./my_dbook --pii
+```
+
+### With LLM enrichment (semantic summaries, concept aliases)
+```bash
+pip install dbook[llm]
+dbook compile "postgresql://..." --output ./my_dbook --llm --llm-provider anthropic --llm-key sk-...
+```
+
+### Check for schema changes
+```bash
+dbook check ./my_dbook "postgresql://user:pass@host/db"
+```
+
+### Incremental recompile (only changed tables)
+```bash
+dbook compile "postgresql://..." --output ./my_dbook --incremental
+```
+
+### Python API
+```python
+from dbook.catalog import SQLAlchemyCatalog
+from dbook.compiler import compile_book
+from dbook.validator import QueryValidator
+
+# Compile
+catalog = SQLAlchemyCatalog("postgresql://user:pass@host/db")
+book = catalog.introspect_all()
+compile_book(book, "./my_dbook")
+
+# Validate agent SQL
+validator = QueryValidator(book)
+result = validator.validate("SELECT * FROM orders WHERE status = 'delivered'")
+print(result.valid, result.errors, result.warnings)
+```
+
+## Optional Features
+
+| Feature | Install | Flag | What it adds |
+|---------|---------|------|-------------|
+| PII detection | `pip install dbook[pii]` | `--pii` | Column sensitivity markers, sample data redaction |
+| LLM enrichment | `pip install dbook[llm]` | `--llm` | Semantic summaries, concept aliases, schema narratives |
+| Metrics | `pip install dbook[metrics]` | `--metrics` | User-defined canonical business metrics |
+
+## The Silver Layer Insight
+
+Traditional data pipelines create gold layers because consumers can't read raw data. With dbook, AI agents can understand silver directly — reducing the need for gold views for discovery and ad-hoc queries.
+
+<p align="center">
+  <img src="docs/silver-layer.svg" alt="The Silver Layer Insight" width="800">
+</p>
+
+> **Note:** dbook reduces the need for gold views for discovery and ad-hoc queries.
+> Gold layers still provide value for: enforced business rules, canonical metric
+> definitions, data quality guarantees, and grain standardization. For critical metrics,
+> define them in `metrics.yaml` — dbook includes them in its output so agents use the
+> canonical definition, not their own interpretation.
+
+## Development
+
+```bash
+pip install -e ".[dev]"
+pytest tests/ -q --tb=short
+```
+
+118 tests covering: introspection, compilation, CLI, PII detection, LLM enrichment, query validation, and realistic agent simulation benchmarks.
+
+## License
+
+Apache License 2.0