minizen 0.1.0__tar.gz

minizen-0.1.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 HieuDao-code
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

minizen-0.1.0/PKG-INFO

@@ -0,0 +1,67 @@
+Metadata-Version: 2.4
+Name: minizen
+Version: 0.1.0
+Summary: A quieter way to stay informed
+Author: Hieu Dao
+Author-email: Hieu Dao <106801182+HieuDao-code@users.noreply.github.com>
+License-Expression: MIT
+License-File: LICENSE
+Classifier: Development Status :: 3 - Alpha
+Classifier: Environment :: Console
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: End Users/Desktop
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3 :: Only
+Classifier: Programming Language :: Python :: 3.14
+Classifier: Topic :: Internet :: WWW/HTTP :: Dynamic Content :: News/Diary
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Requires-Dist: miniflux>=1.1.6,<2.0.0
+Requires-Dist: mistune>=3.2.0,<4.0.0
+Requires-Dist: pydantic>=2.13.1,<3.0.0
+Requires-Dist: pydantic-ai-slim[anthropic,openai]>=1.84.0,<2.0.0
+Requires-Dist: python-dotenv>=1.1.0,<2.0.0
+Requires-Dist: tomli-w>=1.0.0,<2.0.0
+Requires-Dist: typer>=0.24.1,<0.25.0
+Requires-Python: >=3.14
+Project-URL: Homepage, https://github.com/hieudao-code/minizen
+Project-URL: Documentation, https://hieudao-code.github.io/minizen/
+Project-URL: Repository, https://github.com/hieudao-code/minizen
+Project-URL: Issues, https://github.com/hieudao-code/minizen/issues
+Project-URL: Changelog, https://github.com/hieudao-code/minizen/releases
+Description-Content-Type: text/markdown
+
+<h1 align="center">
+  <p align="center">
+    <span style="font-size: 80px; font-weight: bold;">mini</span>
+    <span style="font-size: 80px; font-weight: bold; color: #F27D3E;">zen</span>
+  </p>
+</h1>
+<p align="center">A quieter way to stay informed.</p>
+<p align="center">
+  <a href="https://hieudao-code.github.io/minizen/">Documentation</a> | <a href="https://hieudao-code.github.io/minizen/getting_started/">Getting Started</a> | <a href="https://hieudao-code.github.io/minizen/configuration/">Configuration</a>
+</p>
+
+## About
+
+**minizen** fetches your unread RSS articles from [Miniflux](https://miniflux.app),
+uses AI to curate and summarise the most interesting ones, and emails you a clean daily digest.
+
+### Features
+
+- **Curated, not firehosed** — the AI picks your top N articles and writes a cohesive
+  narrative, not a bullet dump
+- **Runs on a schedule** — ships with a GitHub Actions workflow for a hands-free daily digest,
+  no server required
+- **Dry-run friendly** — preview the digest in your terminal before a single email is sent
+- **Pluggable AI** — works with Anthropic Claude or OpenAI models via
+  [pydantic-ai](https://ai.pydantic.dev/)
+
+## Quick start
+
+```bash
+uv tool install minizen
+minizen setup            # interactive wizard — configure Miniflux, AI, and email
+minizen digest preview   # preview today's digest in your terminal
+minizen run              # fetch → summarise → send
+```

minizen-0.1.0/README.md

@@ -0,0 +1,34 @@
+<h1 align="center">
+  <p align="center">
+    <span style="font-size: 80px; font-weight: bold;">mini</span>
+    <span style="font-size: 80px; font-weight: bold; color: #F27D3E;">zen</span>
+  </p>
+</h1>
+<p align="center">A quieter way to stay informed.</p>
+<p align="center">
+  <a href="https://hieudao-code.github.io/minizen/">Documentation</a> | <a href="https://hieudao-code.github.io/minizen/getting_started/">Getting Started</a> | <a href="https://hieudao-code.github.io/minizen/configuration/">Configuration</a>
+</p>
+
+## About
+
+**minizen** fetches your unread RSS articles from [Miniflux](https://miniflux.app),
+uses AI to curate and summarise the most interesting ones, and emails you a clean daily digest.
+
+### Features
+
+- **Curated, not firehosed** — the AI picks your top N articles and writes a cohesive
+  narrative, not a bullet dump
+- **Runs on a schedule** — ships with a GitHub Actions workflow for a hands-free daily digest,
+  no server required
+- **Dry-run friendly** — preview the digest in your terminal before a single email is sent
+- **Pluggable AI** — works with Anthropic Claude or OpenAI models via
+  [pydantic-ai](https://ai.pydantic.dev/)
+
+## Quick start
+
+```bash
+uv tool install minizen
+minizen setup            # interactive wizard — configure Miniflux, AI, and email
+minizen digest preview   # preview today's digest in your terminal
+minizen run              # fetch → summarise → send
+```

minizen-0.1.0/pyproject.toml

@@ -0,0 +1,168 @@
+[build-system]
+requires = ["uv_build>=0.11.7,<0.12"]
+build-backend = "uv_build"
+
+[project]
+name = "minizen"
+version = "0.1.0"
+description = "A quieter way to stay informed"
+readme = "README.md"
+license = "MIT"
+license-files = ["LICENSE"]
+requires-python = ">=3.14"
+authors = [
+  { name = "Hieu Dao", email = "106801182+HieuDao-code@users.noreply.github.com" },
+]
+classifiers = [
+  "Development Status :: 3 - Alpha",
+  "Environment :: Console",
+  "Intended Audience :: Developers",
+  "Intended Audience :: End Users/Desktop",
+  "Operating System :: OS Independent",
+  "Programming Language :: Python :: 3",
+  "Programming Language :: Python :: 3 :: Only",
+  "Programming Language :: Python :: 3.14",
+  "Topic :: Internet :: WWW/HTTP :: Dynamic Content :: News/Diary",
+  "Topic :: Scientific/Engineering :: Artificial Intelligence",
+]
+dependencies = [
+  "miniflux>=1.1.6,<2.0.0",
+  "mistune>=3.2.0,<4.0.0",
+  "pydantic>=2.13.1,<3.0.0",
+  "pydantic-ai-slim[anthropic,openai]>=1.84.0,<2.0.0",
+  "python-dotenv>=1.1.0,<2.0.0",
+  "tomli-w>=1.0.0,<2.0.0",
+  "typer>=0.24.1,<0.25.0",
+]
+
+[project.urls]
+Homepage = "https://github.com/hieudao-code/minizen"
+Documentation = "https://hieudao-code.github.io/minizen/"
+Repository = "https://github.com/hieudao-code/minizen"
+Issues = "https://github.com/hieudao-code/minizen/issues"
+Changelog = "https://github.com/hieudao-code/minizen/releases"
+
+[project.scripts]
+minizen = "minizen.main:app"
+mz = "minizen.main:app"
+
+[dependency-groups]
+dev = [
+  { include-group = "docs" },
+  { include-group = "lint" },
+  { include-group = "miscellaneous" },
+  { include-group = "release" },
+  { include-group = "security" },
+  { include-group = "test" },
+  { include-group = "typing" },
+]
+docs = ["zensical==0.0.32"]
+lint = ["ruff==0.15.10"]
+miscellaneous = ["prek==0.3.9"]
+release = ["commitizen==4.13.10"]
+security = ["zizmor==1.24.1"]
+test = ["pytest==9.0.3", "pytest-cov==7.1.0", "pytest-mock==3.15.1"]
+typing = ["ty==0.0.29"]
+
+[tool.uv]
+exclude-newer = "1 week"
+
+[tool.commitizen]
+name = "cz_conventional_commits"
+tag_format = "$version"
+version_scheme = "pep440"
+version_provider = "uv"
+update_changelog_on_bump = true
+breaking_change_exclamation_in_title = true
+version_files = ["src/minizen/__init__.py"]
+major_version_zero = true
+
+[tool.ruff]
+target-version = "py314"
+
+[tool.ruff.lint]
+select = [
+  "A",     # flake8-builtins
+  "ASYNC", # flake8-async
+  "B",     # flake8-bugbear
+  "C4",    # flake8-comprehensions
+  "C90",   # mccabe complexity
+  "E",     # pycodestyle error
+  "F",     # pyflakes
+  "FAST",  # fastapi
+  "I",     # isort
+  "N",     # pep8-naming
+  "PERF",  # Perflint
+  "PIE",   # flake8-pie
+  "PLR",   # pylint refactor
+  "PT",    # flake8-pytest-style
+  "PYI",   # flake8-pyi
+  "RUF",   # ruff
+  "S",     # flake8-bandit
+  "SIM",   # flake8-simplify
+  "UP",    # pyupgrade
+  "W",     # pycodestyle warnings
+]
+ignore = [
+  "B904", # raise-without-from-inside-except
+]
+
+[tool.ruff.lint.per-file-ignores]
+"**/tests/**" = [
+  "PLR2004", # magic-value-comparison
+  "PLR0913", # too-many-arguments
+  "S101",    # assert
+  "S105",    # hardcoded-password-string
+  "S106",    # hardcoded-password-func-arg
+]
+"src/minizen/cli/commands/setup.py" = ["PLR0913"]
+"src/minizen/cli/commands/run.py" = ["PLR0913"]
+"src/minizen/providers/email/template.py" = ["E501"]
+
+[tool.ruff.lint.isort]
+known-first-party = ["minizen"]
+force-wrap-aliases = true
+combine-as-imports = true
+
+[tool.ruff.format]
+docstring-code-format = true
+
+[tool.coverage.run]
+branch = true
+source = ["minizen"]
+omit = ["*/minizen/main.py"]
+
+[tool.pytest.ini_options]
+addopts = "--cov --cov-report=term-missing --cov-fail-under=100"
+testpaths = ["tests"]
+
+[tool.tox]
+env_list = ["py314", "commit-msg", "lint", "security", "typing"]
+
+[tool.tox.env_run_base]
+description = "Run tests"
+dependency_groups = ["dev"]
+commands = [["pytest"]]
+
+[tool.tox.env.commit-msg]
+description = "Run commitizen checks"
+skip_install = true
+dependency_groups = ["release"]
+commands = [["cz", "check", "--rev-range", "origin/main..HEAD"]]
+
+[tool.tox.env.lint]
+description = "Run linters"
+skip_install = true
+dependency_groups = ["lint"]
+commands = [["ruff", "check"], ["ruff", "format", "--check"]]
+
+[tool.tox.env.security]
+description = "Run security checks"
+skip_install = true
+dependency_groups = ["security"]
+commands = [["uv", "audit"], ["zizmor", '.', "--persona", "auditor"]]
+
+[tool.tox.env.typing]
+description = "Run type checks"
+dependency_groups = ["test", "typing"]
+commands = [["ty", "check"]]

minizen-0.1.0/src/minizen/__init__.py

@@ -0,0 +1,29 @@
+"""minizen — A quieter way to stay informed."""
+
+from minizen.ai import DigestAgent, DigestResult
+from minizen.config import (
+    AIConfig,
+    EmailConfig,
+    MinifluxConfig,
+    Settings,
+    load_settings,
+)
+from minizen.core import run_pipeline
+from minizen.providers.email import EmailProvider
+from minizen.providers.rss import Article, MinifluxProvider
+
+__version__ = "0.1.0"
+
+__all__ = [
+    "AIConfig",
+    "Article",
+    "DigestAgent",
+    "DigestResult",
+    "EmailConfig",
+    "EmailProvider",
+    "MinifluxConfig",
+    "MinifluxProvider",
+    "Settings",
+    "load_settings",
+    "run_pipeline",
+]

minizen-0.1.0/src/minizen/ai/__init__.py

@@ -0,0 +1,5 @@
+"""Public AI agent API for minizen."""
+
+from minizen.ai.agent import DigestAgent, DigestResult
+
+__all__ = ["DigestAgent", "DigestResult"]

minizen-0.1.0/src/minizen/ai/agent.py

@@ -0,0 +1,74 @@
+import logging
+from typing import cast
+
+from pydantic import BaseModel, Field
+from pydantic_ai import Agent, AgentRunResult
+
+from minizen.providers.rss.miniflux import Article
+
+logger = logging.getLogger(__name__)
+
+_SYSTEM_PROMPT = """\
+You are a personal news curator. You receive a list of unread articles and must:
+1. Select the top N most important and interesting articles.
+2. Write a cohesive, well-structured Markdown digest covering those articles.
+3. Return the digest and the IDs of the articles you selected.
+
+Each article section must include a direct link to the original article URL.
+Be concise. Prioritise articles with broad significance over niche topics.
+"""
+
+
+class DigestResult(BaseModel):
+    """Structured output from the AI digest agent."""
+
+    markdown: str = Field(description="Markdown digest text produced by the agent.")
+    articles_used: list[int] = Field(
+        description="IDs of the articles selected for the digest."
+    )
+
+
+class DigestAgent:
+    """AI-powered agent that selects and summarises articles into a Markdown digest."""
+
+    def __init__(self, *, model: str, top_n: int) -> None:
+        """Initialise the agent with the given model and article limit.
+
+        Args:
+            model: pydantic-ai model identifier (e.g. ``anthropic:claude-haiku-4-5``).
+            top_n: Maximum number of articles to include in the digest.
+        """
+        logger.debug("Initialising DigestAgent: model=%s, top_n=%d", model, top_n)
+        self._top_n = top_n
+        self._agent = Agent(
+            model=model,
+            output_type=DigestResult,
+            system_prompt=_SYSTEM_PROMPT,
+        )
+
+    def run(self, *, articles: list[Article]) -> DigestResult:
+        """Select the top N articles and return a structured Markdown digest.
+
+        Args:
+            articles: Full list of unread articles to choose from.
+
+        Returns:
+            A ``DigestResult`` containing the Markdown text and the IDs of
+            articles that were included.
+        """
+        logger.info("Running AI agent on %d article(s)", len(articles))
+        articles_text = "\n\n---\n\n".join(
+            f"ID: {a.id}\n"
+            f"Feed: {a.feed_name}\n"
+            f"Title: {a.title}\n"
+            f"URL: {a.url}\n"
+            f"Published: {a.published_at.isoformat()}\n\n"
+            f"{a.content}"
+            for a in articles
+        )
+        user_prompt = (
+            f"Please select the top {self._top_n} most important articles "  # noqa: S608
+            f"from the following and write a digest:\n\n{articles_text}"
+        )
+        result = self._agent.run_sync(user_prompt)
+        return cast(AgentRunResult[DigestResult], result).output

minizen-0.1.0/src/minizen/cli/__init__.py

@@ -0,0 +1,34 @@
+from typing import Annotated
+
+import typer
+
+from minizen.cli.commands import (
+    config as config_module,
+    digest as digest_module,
+)
+from minizen.cli.commands.run import run
+from minizen.cli.commands.setup import setup
+from minizen.cli.state import configure_logging
+
+app = typer.Typer(
+    name="minizen",
+    help="A quieter way to stay informed.",
+    no_args_is_help=True,
+)
+
+
+@app.callback()
+def _callback(
+    verbose: Annotated[
+        bool,
+        typer.Option("--verbose", "-v", help="Enable debug logging."),
+    ] = False,
+) -> None:
+    """Configure logging for the CLI session."""
+    configure_logging(verbose=verbose)
+
+
+app.command("run")(run)
+app.command("setup")(setup)
+app.add_typer(config_module.app, name="config")
+app.add_typer(digest_module.app, name="digest")

minizen-0.1.0/src/minizen/cli/commands/config.py

@@ -0,0 +1,105 @@
+import contextlib
+import tomllib
+from pathlib import Path
+from typing import Annotated
+
+import tomli_w
+import typer
+
+from minizen.config.loader import load_settings
+
+_DEFAULT_CONFIG = Path.home() / ".config" / "minizen" / "config.toml"
+
+_CONFIG_OPTION = Annotated[
+    Path,
+    typer.Option(help="Path to the TOML configuration file.", show_default=True),
+]
+
+app = typer.Typer(help="Inspect and update configuration.")
+
+
+@app.command("show")
+def show(config: _CONFIG_OPTION = _DEFAULT_CONFIG) -> None:
+    """Display the current configuration."""
+    try:
+        with open(config, "rb") as f:
+            data = tomllib.load(f)
+    except FileNotFoundError:
+        typer.echo(f"Config file not found: {config}")
+        typer.echo("Run `minizen setup` to create one.")
+        raise typer.Exit(code=1)
+
+    typer.echo(f"Config file: {config}")
+    mf = data.get("miniflux", {})
+    em = data.get("email", {})
+    ai = data.get("ai", {})
+    typer.echo(f"  miniflux.url:       {mf.get('url', '(unset)')}")
+    typer.echo(f"  miniflux.api_key:   {'(from env)' if mf.get('url') else '(unset)'}")
+    typer.echo(f"  email.smtp_host:    {em.get('smtp_host', '(unset)')}")
+    typer.echo(f"  email.smtp_port:    {em.get('smtp_port', '(unset)')}")
+    typer.echo(f"  email.from_addr:    {em.get('from_addr', '(unset)')}")
+    typer.echo(f"  email.to_addr:      {em.get('to_addr', '(unset)')}")
+    typer.echo(f"  ai.model:           {ai.get('model', 'anthropic:claude-haiku-4-5')}")
+    typer.echo(f"  ai.top_n:           {ai.get('top_n', 5)}")
+
+
+@app.command("validate")
+def validate(config: _CONFIG_OPTION = _DEFAULT_CONFIG) -> None:
+    """Validate the configuration file and environment variables."""
+    try:
+        load_settings(config_path=config)
+    except FileNotFoundError:
+        typer.echo(f"Config file not found: {config}")
+        typer.echo("Run `minizen setup` to create one.")
+        raise typer.Exit(code=1)
+    except KeyError as e:
+        typer.echo(f"Error: missing environment variable {e.args[0]}")
+        raise typer.Exit(code=1)
+    typer.echo("Configuration is valid.")
+
+
+_ALLOWED_KEYS = {
+    "miniflux.url",
+    "ai.model",
+    "ai.top_n",
+    "email.smtp_host",
+    "email.smtp_port",
+    "email.from_addr",
+    "email.to_addr",
+}
+
+
+@app.command("set")
+def set_value(
+    key: Annotated[
+        str, typer.Argument(help="Dot-separated config key (e.g. ai.top_n).")
+    ],
+    value: Annotated[str, typer.Argument(help="New value.")],
+    config: _CONFIG_OPTION = _DEFAULT_CONFIG,
+) -> None:
+    """Set a configuration value in the TOML file."""
+    if key not in _ALLOWED_KEYS:
+        allowed = ", ".join(sorted(_ALLOWED_KEYS))
+        typer.echo(f"Error: unknown config key '{key}'. Allowed: {allowed}")
+        raise typer.Exit(code=1)
+
+    try:
+        with open(config, "rb") as f:
+            data = tomllib.load(f)
+    except FileNotFoundError:
+        typer.echo(f"Config file not found: {config}")
+        typer.echo("Run `minizen setup` to create one.")
+        raise typer.Exit(code=1)
+
+    section, field = key.split(".", 1)
+    if section not in data:
+        data[section] = {}
+
+    # Coerce numeric strings to int
+    coerced: str | int = value
+    with contextlib.suppress(ValueError):
+        coerced = int(value)
+
+    data[section][field] = coerced
+    config.write_bytes(tomli_w.dumps(data).encode())
+    typer.echo(f"Set {key} = {coerced!r}")

minizen-0.1.0/src/minizen/cli/commands/digest.py

@@ -0,0 +1,135 @@
+from datetime import date
+from pathlib import Path
+from typing import Annotated
+
+import typer
+
+from minizen.ai.agent import DigestAgent
+from minizen.cli.state import configure_logging
+from minizen.config.loader import load_settings
+from minizen.config.models import Settings
+from minizen.providers.email.smtp import EmailProvider
+from minizen.providers.email.template import render_email
+from minizen.providers.rss.miniflux import MinifluxProvider
+
+_DEFAULT_CONFIG = Path.home() / ".config" / "minizen" / "config.toml"
+
+_CONFIG_OPTION = Annotated[
+    Path,
+    typer.Option(help="Path to the TOML configuration file.", show_default=True),
+]
+
+_VERBOSE_OPTION = Annotated[
+    bool,
+    typer.Option("--verbose", "-v", help="Enable debug logging."),
+]
+
+_DRY_RUN_OPTION = Annotated[
+    bool,
+    typer.Option(
+        "--dry-run",
+        help="Fetch articles but skip LLM call and any external sends.",
+    ),
+]
+
+app = typer.Typer(help="Preview or test the digest without marking articles as read.")
+
+
+def _load(config: Path) -> Settings:
+    """Load settings from a TOML file, exiting with an error message on failure.
+
+    Args:
+        config: Path to the TOML configuration file.
+
+    Returns:
+        Fully loaded application settings.
+    """
+    try:
+        return load_settings(config_path=config)
+    except FileNotFoundError:
+        typer.echo(f"Config file not found: {config}")
+        typer.echo("Run `minizen setup` to create one.")
+        raise typer.Exit(code=1)
+    except KeyError as e:
+        typer.echo(f"Error: missing environment variable {e.args[0]}")
+        raise typer.Exit(code=1)
+
+
+@app.command("fetch")
+def fetch(
+    config: _CONFIG_OPTION = _DEFAULT_CONFIG,
+    verbose: _VERBOSE_OPTION = False,
+) -> None:
+    """Fetch unread articles and print their titles and URLs."""
+    configure_logging(verbose=verbose)
+    settings = _load(config)
+    rss = MinifluxProvider(config=settings.miniflux)
+    articles = rss.fetch_unread()
+    if not articles:
+        typer.echo("No unread articles.")
+        return
+    typer.echo(f"{len(articles)} unread article(s):\n")
+    for article in articles:
+        typer.echo(f"[{article.feed_name}] {article.title}")
+        typer.echo(f"  {article.url}")
+
+
+@app.command("preview")
+def preview(
+    config: _CONFIG_OPTION = _DEFAULT_CONFIG,
+    verbose: _VERBOSE_OPTION = False,
+    dry_run: _DRY_RUN_OPTION = False,
+) -> None:
+    """Fetch and summarise articles, then print the Markdown digest."""
+    configure_logging(verbose=verbose)
+    settings = _load(config)
+    rss = MinifluxProvider(config=settings.miniflux)
+    articles = rss.fetch_unread()
+    if not articles:
+        typer.echo("No unread articles.")
+        return
+    if dry_run:
+        typer.echo(f"{len(articles)} unread article(s):\n")
+        for article in articles:
+            typer.echo(f"[{article.feed_name}] {article.title}")
+            typer.echo(f"  {article.url}")
+        return
+    agent = DigestAgent(model=settings.ai.model, top_n=settings.ai.top_n)
+    result = agent.run(articles=articles)
+    typer.echo(result.markdown)
+
+
+@app.command("send-test")
+def send_test(
+    config: _CONFIG_OPTION = _DEFAULT_CONFIG,
+    verbose: _VERBOSE_OPTION = False,
+    dry_run: _DRY_RUN_OPTION = False,
+) -> None:
+    """Send a test digest email without marking articles as read."""
+    configure_logging(verbose=verbose)
+    settings = _load(config)
+    rss = MinifluxProvider(config=settings.miniflux)
+    articles = rss.fetch_unread()
+    if not articles:
+        typer.echo("No unread articles.")
+        return
+    if dry_run:
+        typer.confirm(
+            "This will make a real LLM API call but will not send an email. Continue?",
+            abort=True,
+        )
+    agent = DigestAgent(model=settings.ai.model, top_n=settings.ai.top_n)
+    result = agent.run(articles=articles)
+    html, plain_text = render_email(result.markdown)
+    if dry_run:
+        typer.echo("Dry run — email not sent:\n")
+        typer.echo(plain_text)
+        return
+    today = date.today().strftime("%B %-d, %Y")
+    email = EmailProvider(config=settings.email)
+    email.send(
+        subject=f"[TEST] Your Daily Zen — {today}",
+        html=html,
+        plain_text=plain_text,
+    )
+    typer.echo("Test digest sent.")