perag 0.1.0__tar.gz

perag-0.1.0/.gitignore

@@ -0,0 +1,221 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[codz]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#   Usually these files are written by a python script from a template
+#   before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py.cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+# Pipfile.lock
+
+# UV
+#   Similar to Pipfile.lock, it is generally recommended to include uv.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+# uv.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+# poetry.lock
+# poetry.toml
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#   pdm recommends including project-wide configuration in pdm.toml, but excluding .pdm-python.
+#   https://pdm-project.org/en/latest/usage/project/#working-with-version-control
+# pdm.lock
+# pdm.toml
+.pdm-python
+.pdm-build/
+
+# pixi
+#   Similar to Pipfile.lock, it is generally recommended to include pixi.lock in version control.
+# pixi.lock
+#   Pixi creates a virtual environment in the .pixi directory, just like venv module creates one
+#   in the .venv directory. It is recommended not to include this directory in version control.
+.pixi
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# Redis
+*.rdb
+*.aof
+*.pid
+
+# RabbitMQ
+mnesia/
+rabbitmq/
+rabbitmq-data/
+
+# ActiveMQ
+activemq-data/
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.envrc
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#   JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#   be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#   and can be added to the global gitignore or merged into this file.  For a more nuclear
+#   option (not recommended) you can uncomment the following to ignore the entire idea folder.
+# .idea/
+
+# Abstra
+#   Abstra is an AI-powered process automation framework.
+#   Ignore directories containing user credentials, local state, and settings.
+#   Learn more at https://abstra.io/docs
+.abstra/
+
+# Visual Studio Code
+#   Visual Studio Code specific template is maintained in a separate VisualStudioCode.gitignore 
+#   that can be found at https://github.com/github/gitignore/blob/main/Global/VisualStudioCode.gitignore
+#   and can be added to the global gitignore or merged into this file. However, if you prefer, 
+#   you could uncomment the following to ignore the entire vscode folder
+# .vscode/
+# Temporary file for partial code execution
+tempCodeRunnerFile.py
+
+# Ruff stuff:
+.ruff_cache/
+
+# PyPI configuration file
+.pypirc
+
+# Marimo
+marimo/_static/
+marimo/_lsp/
+__marimo__/
+
+# Streamlit
+.streamlit/secrets.toml
+/.claude/
+/.perag/
+/.venv/

perag-0.1.0/.idea/.gitignore

@@ -0,0 +1,3 @@
+# Default ignored files
+/shelf/
+/workspace.xml

perag-0.1.0/.idea/encodings.xml

@@ -0,0 +1,4 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="Encoding" addBOMForNewFiles="with NO BOM" />
+</project>

perag-0.1.0/.idea/inspectionProfiles/Project_Default.xml

@@ -0,0 +1,13 @@
+<component name="InspectionProjectProfileManager">
+  <profile version="1.0">
+    <option name="myName" value="Project Default" />
+    <inspection_tool class="PyStubPackagesAdvertiser" enabled="true" level="WARNING" enabled_by_default="true">
+      <option name="ignoredPackages">
+        <list>
+          <option value="pyspark-stubs==3.0.0.dev8" />
+        </list>
+      </option>
+    </inspection_tool>
+    <inspection_tool class="ReassignedToPlainText" enabled="false" level="WARNING" enabled_by_default="false" />
+  </profile>
+</component>

perag-0.1.0/.idea/inspectionProfiles/profiles_settings.xml

@@ -0,0 +1,6 @@
+<component name="InspectionProjectProfileManager">
+  <settings>
+    <option name="USE_PROJECT_PROFILE" value="false" />
+    <version value="1.0" />
+  </settings>
+</component>

perag-0.1.0/.idea/misc.xml

@@ -0,0 +1,7 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="Black">
+    <option name="sdkName" value="Python 3.13 virtualenv at ~/github/pet/.venv" />
+  </component>
+  <component name="ProjectRootManager" version="2" project-jdk-name="Python 3.13 virtualenv at ~/github/pet/.venv" project-jdk-type="Python SDK" />
+</project>

perag-0.1.0/.idea/modules.xml

@@ -0,0 +1,8 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="ProjectModuleManager">
+    <modules>
+      <module fileurl="file://$PROJECT_DIR$/.idea/perag.iml" filepath="$PROJECT_DIR$/.idea/perag.iml" />
+    </modules>
+  </component>
+</project>

perag-0.1.0/.idea/perag.iml

@@ -0,0 +1,8 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<module type="PYTHON_MODULE" version="4">
+  <component name="NewModuleRootManager">
+    <content url="file://$MODULE_DIR$" />
+    <orderEntry type="jdk" jdkName="Python 3.13 virtualenv at ~/github/pet/.venv" jdkType="Python SDK" />
+    <orderEntry type="sourceFolder" forTests="false" />
+  </component>
+</module>

perag-0.1.0/.idea/vcs.xml

@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="VcsDirectoryMappings">
+    <mapping directory="$PROJECT_DIR$" vcs="Git" />
+  </component>
+</project>

perag-0.1.0/.idea/workspace.xml

@@ -0,0 +1,82 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="AutoImportSettings">
+    <option name="autoReloadType" value="SELECTIVE" />
+  </component>
+  <component name="ChangeListManager">
+    <list default="true" id="662c9670-5acf-45d8-9761-ba46d1f48197" name="Changes" comment="">
+      <change afterPath="$PROJECT_DIR$/.python-version" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/CLAUDE.md" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/chunkers/__init__.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/chunkers/base.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/chunkers/docx.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/chunkers/markdown.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/chunkers/pdf.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/chunkers/registry.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/chunkers/text.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/config.example.toml" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/db/__init__.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/db/search.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/db/store.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/docs/chunking.md" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/docs/embedders.md" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/docs/pipeline.md" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/embedders/__init__.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/embedders/base.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/embedders/local.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/embedders/ollama.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/embedders/openai.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/embedders/registry.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/perag/__init__.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/perag/cli.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/perag/config.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/perag/schema.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/pyproject.toml" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/skills/SKILL.md" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/uv.lock" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/.gitignore" beforeDir="false" afterPath="$PROJECT_DIR$/.gitignore" afterDir="false" />
+    </list>
+    <option name="SHOW_DIALOG" value="false" />
+    <option name="HIGHLIGHT_CONFLICTS" value="true" />
+    <option name="HIGHLIGHT_NON_ACTIVE_CHANGELIST" value="false" />
+    <option name="LAST_RESOLUTION" value="IGNORE" />
+  </component>
+  <component name="Git.Settings">
+    <option name="RECENT_GIT_ROOT_PATH" value="$PROJECT_DIR$" />
+  </component>
+  <component name="ProjectColorInfo"><![CDATA[{
+  "associatedIndex": 3
+}]]></component>
+  <component name="ProjectId" id="3EDonFdWD89jDVglvrZjum7PGi0" />
+  <component name="ProjectViewState">
+    <option name="hideEmptyMiddlePackages" value="true" />
+    <option name="showLibraryContents" value="true" />
+  </component>
+  <component name="PropertiesComponent"><![CDATA[{
+  "keyToString": {
+    "ModuleVcsDetector.initialDetectionPerformed": "true",
+    "RunOnceActivity.ShowReadmeOnStart": "true",
+    "git-widget-placeholder": "main",
+    "junie.onboarding.icon.badge.shown": "true",
+    "last_opened_file_path": "/Users/verhasp/github/perag",
+    "settings.editor.selected.configurable": "preferences.pluginManager"
+  }
+}]]></component>
+  <component name="SharedIndexes">
+    <attachedChunks>
+      <set>
+        <option value="bundled-python-sdk-9f8e2b94138c-36ea0e71a18c-com.jetbrains.pycharm.pro.sharedIndexes.bundled-PY-251.26094.141" />
+      </set>
+    </attachedChunks>
+  </component>
+  <component name="TaskManager">
+    <task active="true" id="Default" summary="Default task">
+      <changelist id="662c9670-5acf-45d8-9761-ba46d1f48197" name="Changes" comment="" />
+      <created>1779722665327</created>
+      <option name="number" value="Default" />
+      <option name="presentableId" value="Default" />
+      <updated>1779722665327</updated>
+    </task>
+    <servers />
+  </component>
+</project>

perag-0.1.0/.python-version

@@ -0,0 +1 @@
+3.14

perag-0.1.0/CLAUDE.md

@@ -0,0 +1,312 @@
+# Perago — Personal RAG Toolkit
+
+## Project Overview
+
+**Perago** (Latin: *to carry through to completion*) is a personal productivity tool for
+non-developers who work with textual documents (PDF, Word, Markdown, plain text). It
+provides a local, private RAG (Retrieval-Augmented Generation) pipeline that enriches
+prompts with relevant context retrieved from a personal document collection.
+
+The command-line tool is called `perag`. It is intentionally not a service — no server,
+no daemon, no cloud dependency. It runs locally, stores data locally, and is invoked
+from the command line or by Claude Code via a SKILL.md.
+
+---
+
+## Repository Structure
+
+```
+perago/
+├── CLAUDE.md                  # This file
+├── README.md                  # User-facing documentation
+├── pyproject.toml             # Root package definition (if monorepo build)
+├── config.example.toml        # Example configuration file
+│
+├── perag/                     # Main CLI package (entry point: perag <subcommand>)
+│   ├── __init__.py
+│   ├── cli.py                 # Subcommand dispatcher (chunk/embed/ingest/query)
+│   ├── config.py              # Config loading from config.toml
+│   └── schema.py              # Shared JSON chunk schema (dataclass/TypedDict)
+│
+├── chunkers/                  # Format-aware chunkers (one module per format)
+│   ├── __init__.py
+│   ├── base.py                # Abstract base class: Chunker.chunk(path) -> [Chunk]
+│   ├── pdf.py                 # PDF chunking via pdfplumber
+│   ├── docx.py                # Word chunking via python-docx
+│   ├── markdown.py            # Markdown chunking via markdown-it-py
+│   ├── text.py                # Plain text / paragraph-aware fallback
+│   └── registry.py            # Maps file extension -> Chunker class
+│
+├── embedders/                 # Embedding providers (one module per provider)
+│   ├── __init__.py
+│   ├── base.py                # Abstract base class: Embedder.embed([str]) -> [[float]]
+│   ├── ollama.py              # Ollama HTTP API
+│   ├── openai.py              # OpenAI embeddings API
+│   ├── local.py               # sentence-transformers (fully local, no API key)
+│   └── registry.py            # Maps provider name -> Embedder class
+│
+├── db/                        # sqlite-vec database layer
+│   ├── __init__.py
+│   ├── store.py               # Schema init, upsert, meta table management
+│   └── search.py              # ANN query, returns top-k chunks
+│
+├── tests/
+│   ├── fixtures/              # Sample PDF, DOCX, MD, TXT files for testing
+│   ├── test_chunkers.py
+│   ├── test_embedders.py
+│   ├── test_db.py
+│   └── test_pipeline.py       # End-to-end: chunk -> embed -> ingest -> query
+│
+├── skills/
+│   └── SKILL.md               # Claude Code skill: how to use perag from Claude Code
+│
+└── docs/
+    ├── chunking.md            # How chunking works per format
+    ├── embedders.md           # Supported embedding providers and configuration
+    └── pipeline.md            # Full pipeline walkthrough
+```
+
+---
+
+## Subcommand Design
+
+All subcommands read/write JSON on stdin/stdout, making the pipeline composable:
+
+```bash
+# Full pipeline (piped)
+perag chunk document.pdf | perag embed | perag ingest
+
+# Full pipeline (with intermediate files for inspection/debugging)
+perag chunk document.pdf      > chunks.json
+perag embed   < chunks.json   > chunks_embedded.json
+perag ingest  < chunks_embedded.json
+
+# Query
+perag query "what are the termination conditions?"
+```
+
+### `perag chunk <file>`
+- Detects format from extension
+- Dispatches to the appropriate chunker in `chunkers/`
+- Outputs a JSON array of Chunk objects to stdout
+
+### `perag embed`
+- Reads JSON array of Chunk objects from stdin
+- Calls the configured embedding provider in batches
+- Outputs the same JSON array with `vector` field added
+
+### `perag ingest`
+- Reads JSON array of embedded Chunk objects from stdin
+- Writes to the sqlite-vec database
+- Enforces dimension and model name consistency via the `meta` table
+- Upserts by `id` (re-ingesting an updated document replaces existing chunks)
+
+### `perag init`
+- Creates `.perag/` in the current directory
+- Writes a minimal `config.toml` inheriting from `~/.perag/config.toml` if it exists
+- Adds `.perag/perag.db` to `.gitignore` if a `.gitignore` is present
+- Safe to re-run — never overwrites an existing config
+
+### `perag query "<text>"`
+- Embeds the query text using the configured provider
+- Performs ANN search against the sqlite-vec database
+- Outputs top-k chunks as plain text (suitable for Claude Code context injection)
+- `--json` flag outputs structured JSON instead
+
+---
+
+## JSON Chunk Schema
+
+Every chunk flowing through the pipeline conforms to this schema:
+
+```json
+{
+  "id":                 "contracts/nda_2024.pdf::chunk::7",
+  "source":             "contracts/nda_2024.pdf",
+  "content":            "The agreement shall terminate upon 30 days written notice...",
+  "metadata": {
+    "format":           "pdf",
+    "page":             3,
+    "section":          "Termination"
+  },
+  "embedding_model":    "nomic-embed-text",
+  "embedding_provider": "ollama",
+  "vector":             [0.021, -0.134, 0.087, "..."]
+}
+```
+
+After `perag chunk`, the fields `embedding_model`, `embedding_provider`, and `vector`
+are all `null`. After `perag embed` all three are populated by the embedder with its
+own identity. The `metadata` fields are format-specific and optional for downstream
+consumers.
+
+### Embedder behaviour with pre-embedded chunks
+
+`perag embed` inspects `embedding_model` on every incoming chunk before deciding what
+to do:
+
+| `embedding_model` in chunk | Matches current config? | Action |
+|---|---|---|
+| `null` | — | Embed, populate all three fields |
+| set | yes | Skip — pass through unchanged |
+| set | no | Re-embed, overwrite vector and embedding fields |
+
+This means re-running `perag embed` after changing providers is safe and correct —
+leftover JSON files from a previous run are detected and re-embedded automatically.
+
+### Ingestor validation
+
+The ingestor enforces consistency between the chunk and the database before writing:
+
+| Chunk state | Action |
+|---|---|
+| `vector` is `null` | Hard error: *"chunks have no vectors — run `perag embed` first"* |
+| `embedding_model` matches `meta` table | Ingest (upsert by `id`) |
+| `embedding_model` differs from `meta` table | Hard error: *"embedding model mismatch — re-run `perag embed` or rebuild the database"* |
+
+---
+
+## Database Design
+
+A single SQLite file (`perag.db`) located in `.perag/` in the current directory, or
+falling back to `~/.perag/perag.db` if no local `.perag/` exists.
+
+Tables:
+- **`chunks`**: id, source, content, metadata (JSON), vector (sqlite-vec column)
+- **`meta`**: embedding model name, embedding provider, vector dimensions, creation timestamp
+
+On first `perag ingest`, the meta table is written. On subsequent ingests, both
+`embedding_model` and `embedding_provider` are validated against the meta table — a
+mismatch on either is a hard error with a clear message directing the user to re-embed
+or rebuild the database.
+
+---
+
+## Configuration
+
+Perago uses a local-first lookup strategy, the same pattern as `.git` and `.claude`.
+The tool always checks the current directory first and falls back to the user-level
+global config.
+
+### Lookup order
+
+```
+./.perag/config.toml       # project-local config (may be committed)
+./.perag/perag.db          # project-local database
+~/.perag/config.toml       # user-level defaults (fallback)
+~/.perag/perag.db          # user-level database (fallback)
+```
+
+A researcher with three document collections simply has three directories, each with
+its own `.perag/`. The `cd` is the context switch — no flags, no project names.
+
+The global `~/.perag/config.toml` holds the user's preferred embedding provider and
+model so every new project inherits sensible defaults without repeating configuration.
+A project-local config only needs to override what differs from the global defaults.
+
+### `.gitignore` recommendation
+
+```
+.perag/perag.db        # large, machine-generated — never commit
+.perag/config.toml     # optional: commit if you want to share project config
+```
+
+### Config file format
+
+`~/.perag/config.toml` (user-level defaults):
+
+```toml
+[embedding]
+provider   = "ollama"          # ollama | openai | local
+model      = "nomic-embed-text"
+url        = "http://localhost:11434"   # only for ollama
+# api_key  = "sk-..."                  # only for openai
+batch_size = 32
+
+[query]
+top_k  = 5
+output = "text"                # text | json
+```
+
+`./.perag/config.toml` (project-local override — only specify what differs):
+
+```toml
+[embedding]
+model = "mxbai-embed-large"    # override model for this project only
+```
+
+---
+
+## Development Conventions
+
+### Python version
+Python 3.11+. No older versions. Use `match` statements freely.
+
+### Dependencies
+Managed with `uv`. Lock file committed. No unpinned dependencies in production code.
+
+```
+pdfplumber        # PDF parsing
+python-docx       # Word parsing
+markdown-it-py    # Markdown parsing
+sqlite-vec        # Vector search SQLite extension
+sentence-transformers  # Local embeddings (optional)
+httpx             # HTTP client for Ollama/OpenAI
+tomllib           # Config parsing (stdlib in 3.11+)
+typer             # CLI framework
+rich              # Terminal output formatting
+```
+
+### Code style
+- `ruff` for linting and formatting
+- Type annotations on all public functions
+- Docstrings on all public classes and methods
+- No global state — everything flows through config and explicit arguments
+
+### Testing
+- `pytest` with fixtures in `tests/fixtures/`
+- Each chunker must have at least one real-file test (not mocked)
+- The embedder tests mock the HTTP/model calls — no network in CI
+- One end-to-end pipeline test using the `local` embedder (no API key needed)
+
+### Adding a new chunker
+1. Create `chunkers/<format>.py` implementing `base.Chunker`
+2. Register it in `chunkers/registry.py`
+3. Add at least one fixture file to `tests/fixtures/`
+4. Add tests in `tests/test_chunkers.py`
+5. Document the chunking strategy in `docs/chunking.md`
+
+### Adding a new embedder
+1. Create `embedders/<provider>.py` implementing `base.Embedder`
+2. Register it in `embedders/registry.py`
+3. Add config documentation in `docs/embedders.md`
+
+---
+
+## Claude Code Integration (SKILL.md)
+
+The `skills/SKILL.md` file teaches Claude Code how to use `perag` as a context
+enrichment tool. The typical workflow Claude Code should follow:
+
+1. When given a task that might benefit from document context, run:
+   `perag query "<relevant aspect of the task>"` 
+2. Prepend the output to the working context before responding
+3. When new documents are provided, run the full pipeline:
+   `perag chunk <file> | perag embed | perag ingest`
+
+The SKILL.md lives inside the repo so it is versioned alongside the tool itself.
+Users copy or symlink it into their Claude Code skills directory.
+
+---
+
+## Non-Goals
+
+- No web UI
+- No REST API or daemon mode
+- No multi-user support
+- No cloud sync
+- No support for source code files (use RustRAG or similar for that)
+- No streaming ingestion of live data sources
+
+This is a tool for a person with a folder of documents who wants to ask questions
+across them. It should stay that simple.