Python Substack

An unofficial Python SDK and CLI for managing Substack publications and drafts.

Features

• Inspect authentication and publication status from the terminal.
• List publications and inspect, schedule, publish, or delete drafts.
• Use stable JSON output in scripts and automation.
• Create drafts and publish posts from Python.
• Convert Markdown into Substack's editor document format.
• Upload local images while rendering Markdown.
• Set audience, comment permissions, SEO title, SEO description, slug, sections, and tags.
• Publish now, schedule drafts, or keep drafts unpublished by default.
• Authenticate with email/password, cookies JSON, or a browser cookie string.
• Run a FastMCP server for AI-assisted publishing workflows.

Installation

pip install python-substack

Install the MCP server extra:

pip install "python-substack[mcp]"

Setup

Copy .env.example to .env and fill in one authentication method:

EMAIL=
PASSWORD=
PUBLICATION_URL=
COOKIES_PATH=
COOKIES_STRING=

Use either EMAIL and PASSWORD, or cookie-based authentication with COOKIES_PATH or COOKIES_STRING. Cookie authentication is usually the better option if Substack prompts for captcha or magic-link sign-in.

Newer Substack accounts may only have magic-link sign-in enabled. To set a password, sign out of Substack, choose "Sign in with password", then choose "Set a new password".

CLI Operations

Create a draft from Markdown without publishing it:

substack drafts create post.md

Set metadata and repeat --tag to attach multiple tags:

substack --json drafts create post.md \
  --title "My Post" \
  --subtitle "Optional subtitle" \
  --tag python \
  --tag substack \
  --slug my-post \
  --search-engine-title "SEO title" \
  --search-engine-description "SEO description"

Creation and publishing are intentionally separate. Use the returned draft ID when the draft is ready:

substack drafts create post.md
substack drafts publish 12345 --no-send

Check authentication, the selected publication, and subscriber count:

substack status

List available publications or target one without changing .env:

substack publications list
substack --publication-url https://example.substack.com drafts list

List and inspect drafts:

substack drafts list --limit 10
substack drafts get 12345

Schedule with a timezone-aware ISO 8601 timestamp, or remove a schedule:

substack drafts schedule 12345 --at 2026-08-01T09:00:00+03:00
substack drafts unschedule 12345

Publishing and deletion prompt for confirmation. Use --yes for intentional non-interactive execution:

substack drafts publish 12345 --no-send
substack drafts delete 12345 --yes

Global options must appear before the command. --json returns stable envelopes containing the raw Substack responses:

substack --json drafts list
substack --cookies cookies.json --json status

Quickstart

import os

from dotenv import load_dotenv
from substack import Api

load_dotenv()

api = Api(
    email=os.getenv("EMAIL"),
    password=os.getenv("PASSWORD"),
    publication_url=os.getenv("PUBLICATION_URL"),
)

result = api.create_draft_from_markdown(
    title="Shipping with Python",
    subtitle="A short note from a script",
    markdown="""
# Hello

This draft was created from **Markdown**.

![Alt text](https://example.com/image.png "Image caption")
""",
    tags=["python", "automation"],
    slug="shipping-with-python",
)

print(result["draft"]["id"])

create_draft_from_markdown creates a draft by default. It only publishes when publish=True is passed.

Legacy Content Publishing CLI

The existing standalone commands remain supported for compatibility.

Check authentication without creating a draft:

substack-auth-check

With a cookies JSON file:

substack-auth-check --cookies cookies.json

Publish a Markdown file as a draft:

substack-publish-markdown post.md --title "My Post"

Create and publish:

substack-publish-markdown post.md --title "My Post" --publish

Publish from YAML:

substack-publish-yaml draft.yaml

Useful options:

substack-publish-markdown post.md \
  --title "My Post" \
  --subtitle "Optional subtitle" \
  --tag python \
  --tag substack \
  --slug my-post \
  --search-engine-title "SEO title" \
  --search-engine-description "SEO description"

Cookie Authentication

Cookie authentication avoids logging in with email/password on every run and helps when Substack requires captcha or magic-link sign-in.

Use a cookies JSON file:

import os

from dotenv import load_dotenv
from substack import Api

load_dotenv()

api = Api(
    cookies_path=os.getenv("COOKIES_PATH"),
    publication_url=os.getenv("PUBLICATION_URL"),
)

Or paste a browser cookie header into COOKIES_STRING:

import os

from dotenv import load_dotenv
from substack import Api

load_dotenv()

api = Api(
    cookies_string=os.getenv("COOKIES_STRING"),
    publication_url=os.getenv("PUBLICATION_URL"),
)

To get a cookie string:

1. Sign in to Substack in your browser.
2. Open developer tools.
3. Go to the network tab and refresh Substack.
4. Select a request such as subscription/unred/subscriptions.
5. Copy the full cookie request header value into COOKIES_STRING.

To export a working session to a cookies JSON file:

api.export_cookies("cookies.json")

Then set:

COOKIES_PATH=cookies.json

The CLI also accepts a cookie JSON path:

substack-publish-markdown post.md --cookies cookies.json

Low-Level Post Builder

import os

from dotenv import load_dotenv
from substack import Api
from substack.post import Post

load_dotenv()

api = Api(
    email=os.getenv("EMAIL"),
    password=os.getenv("PASSWORD"),
    publication_url=os.getenv("PUBLICATION_URL"),
)

user_id = api.get_user_id()

post = Post(
    title="How to publish a Substack post using Python",
    subtitle="Created with python-substack",
    user_id=user_id,
    audience="everyone",
    write_comment_permissions="everyone",
)

post.paragraph("This is a paragraph.")
post.add(
    {
        "type": "paragraph",
        "content": [
            {"content": "A link to "},
            {
                "content": "Substack",
                "marks": [{"type": "link", "href": "https://substack.com"}],
            },
        ],
    }
)
post.add({"type": "paywall"})
post.add({"type": "captionedImage", "src": "https://example.com/image.png"})

draft = api.post_draft(post.get_draft())
api.prepublish_draft(draft.get("id"))
api.publish_draft(draft.get("id"))

Markdown Support

from substack.post import Post

post = Post("Title", "Subtitle", user_id=1)
post.from_markdown(
    """
# Heading

Paragraph with **bold**, *italic*, `code`, [links](https://example.com), and footnotes.[^1]

- Lists
- Images

![Alt](local-image.png "Caption")

[^1]: Footnote text.
"""
)

Supported Markdown includes headings, paragraphs, bold, italic, inline code, strikethrough, superscript, subscript, links, images, linked images, image captions, code blocks, blockquotes, ordered lists, unordered lists, horizontal rules, footnotes, LaTeX math, pull quotes, and callouts. See docs/markdown.md for the full reference with examples.

When an Api instance is passed to from_markdown, local image paths are uploaded before the draft is created:

post.from_markdown(markdown_content, api=api)

YAML Drafts

title: "My Post Title"
subtitle: "My Post Subtitle"
audience: "everyone"
write_comment_permissions: "everyone"
search_engine_title: "SEO title"
search_engine_description: "SEO description"
slug: "my-post-title"
tags:
  - python
  - substack
markdown: |
  # Introduction

  This post body is Markdown.

The lower-level node format is also supported:

title: "My Post Title"
subtitle: "My Post Subtitle"
body:
  0:
    type: "heading"
    level: 1
    content: "Introduction"
  1:
    type: "paragraph"
    content: "This is a paragraph."
  2:
    type: "captionedImage"
    src: "local_image.jpg"

MCP Server

Install the MCP extra:

pip install "python-substack[mcp]"

Run the server over stdio:

substack-mcp

Equivalent Python entry point:

python -c "from substack_mcp.mcp_server import main; main()"

Available tools:

• post_draft_from_markdown(...)
• put_draft(draft_id, update_payload)
• add_tags(draft_id, tags)
• prepublish_draft(draft_id)
• publish_draft(draft_id, send=True, share_automatically=False)

Development

pip install pre-commit
pre-commit install
pytest

Run the offline suite with:

pytest -m "not live"

Live Substack tests are not part of normal CI. They are opt-in and require configured credentials:

RUN_SUBSTACK_E2E=1 pytest -m live

The CLI operations smoke tests are separately opt-in. They create, inspect, and delete disposable drafts but never publish them:

RUN_SUBSTACK_CLI_E2E=1 pytest -m live tests/substack/test_cli_end_to_end.py

Release changes are tracked in CHANGELOG.md. The maintainer release process is documented in docs/releasing.md.

Disclaimer

This project is not affiliated with Substack.