Skip to content

README.md

Latest commit

 

History

History
494 lines (366 loc) · 28.3 KB

README.md

File metadata and controls

494 lines (366 loc) · 28.3 KB
Raw

Substack API and MCP server, self-hosted

A Substack API and a Substack MCP server you run on your own computer, in Python, on the session of your own Google Chrome. All 50 Substack tools we run in production at Cotlet: posts and drafts from Markdown, publishing and scheduling, Notes, comments, subscribers, stats, recommendations and direct messages.

Substack has no public API to write posts, publish Notes or manage subscribers, and no official MCP server. This repo gives you both:

  • MCP server for Claude Code, Claude Desktop, Cursor, Codex and any MCP client, over stdio or HTTP. Setup
  • REST API on http://127.0.0.1:8787/api/v1/substack, with OpenAPI docs at /docs. Setup
  • Python: call("list_posts", status="draft"), plus 14 ready scripts in examples/.

It is for technical people who want to run it themselves, for personal use only (license). If you want it running 24/7 on a server, kept working when Substack changes its site, for a team or a business, use the hosted version at cotlet.dev.

Substack MCP server Python 3.11+ License: personal use

git clone https://github.com/cotletdev/substack-api-mcp-server.git && cd substack-api-mcp-server
python3 -m venv .venv && . .venv/bin/activate && pip install -e .
substack-mcp connect                               # reads your Substack session from Chrome
substack-mcp call list_posts
claude mcp add substack -- "$PWD/.venv/bin/substack-mcp" mcp

Contents

How it works

AI agent (MCP) / script (REST, Python) ──▶ substack-mcp on your computer ──your Substack cookies──▶ substack.com
                                                     ▲
                         substack-mcp connect ───────┘ reads them once from your Chrome (DevTools Protocol)
  1. You sign in to Substack in your own Chrome, the way you always do.
  2. substack-mcp connect attaches to that Chrome through the Chrome DevTools Protocol, reads the substack.com cookies and saves them to ~/.config/substack-api-mcp/session.json, readable by your user only. Substack signs in by an emailed code or link, so there is no password to see or store.
  3. Every tool call is a request to Substack's own web app endpoints (substack.com/api/v1 for the reader side, yourname.substack.com/api/v1 for your publication's dashboard) with those cookies, sent with curl_cffi, which copies Chrome's TLS fingerprint. No headless browser runs.

Substack's session cookie (substack.sid) lasts about three months. When Substack signs you out, the server reads the cookies from Chrome again, if Chrome is open with remote debugging on.

Requirements

  • Python 3.11 or newer.
  • Google Chrome 144 or newer, signed in to Substack. Any Chrome or Chromium works with a debugging port.
  • macOS, Linux or Windows.
  • A Substack account. Publisher tools (posts, subscribers, stats, comments) work on the first publication your account runs as admin, owner, editor or contributor. Reader tools (feed, search, Notes, messages) work on any account.

Install

git clone https://github.com/cotletdev/substack-api-mcp-server.git
cd substack-api-mcp-server
python3 -m venv .venv
. .venv/bin/activate              # Windows: .venv\Scripts\activate
pip install -e .
substack-mcp --version

With uv: uv venv && uv pip install -e ., then uv run substack-mcp ....

Setting it up with an AI agent (Claude Code, Codex, Cursor)? Open it in this folder and ask it to set up the repo: AGENTS.md tells it every step, and it asks you only for the steps in Chrome.

To update: git pull && pip install -e .. We copy every change of the hosted Substack tools into this repo.

Connect your Chrome

Option A: your everyday Chrome (recommended)

  1. Open chrome://inspect/#remote-debugging in Chrome and switch remote debugging on. Chrome then accepts DevTools connections from programs on this computer, and asks you before each one.

  2. Make sure you are signed in at substack.com.

  3. Run:

    substack-mcp connect

  4. Chrome asks whether to allow the connection. Click Allow.

Saved 19 Substack cookies to /Users/you/.config/substack-api-mcp/session.json
Signed in to Substack as Maya Chen (@mayachen)

You can switch remote debugging off again afterwards. Leave it on if you want the server to pick up a new session by itself after Substack signs you out.

connect looks for Chrome's DevToolsActivePort file in the default profile folder. For Chrome Beta, Canary or a custom profile folder, pass --profile-dir "<folder>".

Option B: a Chrome with a debugging port

Chrome does not open a debugging port on your default profile, so start a separate one, sign in to Substack there once, and connect to it:

# macOS
"/Applications/Google Chrome.app/Contents/MacOS/Google Chrome" --remote-debugging-port=9222 --user-data-dir="$HOME/.chrome-substack"
# Linux
google-chrome --remote-debugging-port=9222 --user-data-dir="$HOME/.chrome-substack"
# Windows
"C:\Program Files\Google\Chrome\Application\chrome.exe" --remote-debugging-port=9222 --user-data-dir="%USERPROFILE%\.chrome-substack"

substack-mcp connect --cdp http://127.0.0.1:9222

--cdp also takes a ws:// DevTools URL, for example of a Chrome on another machine reached over an SSH tunnel.

Option C: paste the cookie by hand

For a server with no Chrome: in Chrome open DevTools on substack.com, Application > Cookies > https://substack.com, copy the value of substack.sid, then:

substack-mcp connect --token          # paste it, input is hidden
# or, without a session file:
export SUBSTACK_AUTH_TOKEN=...

A pasted token is not refreshed on its own. Paste a new one when Substack signs you out.

Check the session at any time:

substack-mcp status

MCP server

The MCP server lists every tool with its input schema, a description written for agents, and hints: read tools are marked read-only, and tools that publish, message or delete are marked so your client asks before running them.

Claude Code

claude mcp add substack -- /path/to/substack-api-mcp-server/.venv/bin/substack-mcp mcp
# read only:
claude mcp add substack -- /path/to/substack-api-mcp-server/.venv/bin/substack-mcp mcp --read-only

Claude Desktop

~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "substack": {
      "command": "/path/to/substack-api-mcp-server/.venv/bin/substack-mcp",
      "args": ["mcp"]
    }
  }
}

Cursor

~/.cursor/mcp.json, the same block as Claude Desktop. A ready copy is in mcp.json.

Codex

~/.codex/config.toml:

[mcp_servers.substack]
command = "/path/to/substack-api-mcp-server/.venv/bin/substack-mcp"
args = ["mcp"]

Over HTTP

substack-mcp serve also serves MCP over streamable HTTP at http://127.0.0.1:8787/mcp, for clients that take a URL:

substack-mcp serve
claude mcp add --transport http substack http://127.0.0.1:8787/mcp

Then ask your agent things like:

Turn essay.md into a draft for everyone and schedule it for Tuesday 8:00 New York time. Once it is out, post a Note with a link card to it.

Find comments on my last three posts that I have not answered and draft a short reply to each.

Last week's numbers: new and paid subscribers, cancellations, revenue, and opens for each post.

Where did my readers come from in the last 30 days?

REST API

substack-mcp serve                    # http://127.0.0.1:8787, MCP at /mcp, docs at /docs

Each tool has its own route under /api/v1/substack, the same paths as the hosted API at cotlet.dev, so code written against one works with the other. Path parts and the query string are arguments, and a JSON body adds the rest.

curl 'http://127.0.0.1:8787/api/v1/substack/posts?status=scheduled'
curl 'http://127.0.0.1:8787/api/v1/substack/stats?period=30d'
curl -X POST http://127.0.0.1:8787/api/v1/substack/drafts \
  -H 'Content-Type: application/json' \
  -d '{"title": "Notes on pricing", "body_markdown": "## Why I raised prices\n\nThree reasons..."}'

Any tool by name, body as its arguments:

curl -X POST http://127.0.0.1:8787/api/v1/substack/tools/list_comments -d '{"post_id": "148213977"}'
  • GET /api/v1/substack/tools: every tool with its route, input schema and an example answer.
  • GET /docs: interactive OpenAPI docs, GET /openapi.json the schema, for n8n, Postman or a generated client.
  • Answers are JSON. Lists return next_cursor; pass it back as cursor for the next page.

The server listens on 127.0.0.1 only. To reach it from another device of yours, set a token, and every request then needs Authorization: Bearer <token>:

SUBSTACK_MCP_TOKEN=$(openssl rand -hex 24) substack-mcp serve --host 0.0.0.0

Python and the example scripts

Straight from Python, no server needed:

from substack_api_mcp import call
from substack_api_mcp.scripting import paginate, write_rows

posts = list(paginate("list_posts", {"status": "published"}, "posts"))
write_rows(posts, "posts.csv")
draft = call("create_draft", title="Notes on pricing", body_markdown="Three reasons...")
call("schedule_post", post_id=draft["post_id"], publish_at="2026-11-03T08:00:00-05:00")

Or one tool from the shell: substack-mcp call get_post_stats post_id=148213977, or with a JSON object.

Ready scripts in examples/. Each prints JSON, takes --out file.csv or --out file.json where it makes sense, and --help:

Script What it does
draft_from_markdown.py A .md file to a draft: the first heading is the title, headings, links and lists are kept. --publish-at schedules it.
schedule_post.py Schedules a draft for a set time, or --cancel takes it back to drafts.
publish_note.py A Note now or at a set time, --link-latest adds a link card to your newest post.
list_posts.py Published, draft or scheduled posts: title, audience, dates, URL.
get_post.py One post saved as a Markdown file.
weekly_report.py New and paid subscribers, cancellations, revenue, and each new post's opens and new subscribers.
post_stats.py Opens, open rate, views, likes, restacks and new subscribers for your last N posts.
daily_stats.py Subscribers, paid subscribers, followers and views day by day.
traffic_sources.py Where your visitors came from, most views first.
unanswered_comments.py Comments on your last N posts that you have not replied to.
reply_to_comment.py A reply under one comment.
list_subscribers.py Subscribers by plan or join date, with activity rating.
export_subscribers.py The full list as the CSV that Export gives in the dashboard.
add_subscribers.py Adds emails from a text or CSV file as free subscribers, after asking you. 
python examples/draft_from_markdown.py essay.md --publish-at 2026-11-03T08:00:00-05:00
python examples/unanswered_comments.py --last 3 --out todo.csv

On a schedule, cron on your own computer works while it is on: 0 8 * * 1 cd ~/substack-api-mcp-server && .venv/bin/python examples/weekly_report.py.

All 50 tools

post_id comes from list_posts, comment_id from list_comments. substack-mcp tools prints this list, substack-mcp tools --json adds the input schemas.

Posts and drafts

Tool Access REST route What it does
list_posts read GET /posts Posts of your publication, newest first, with status and audience.
get_post read GET /posts/{post_id} Full post: title, subtitle, body as Markdown, audience, URL and dates.
create_draft write POST /drafts New draft from Markdown. Nothing is sent until you publish or schedule it.
update_draft write PATCH /drafts/{post_id} Change the title, subtitle, body or audience of a draft.
duplicate_post write POST /posts/{post_id}/duplicate Copy any post into a new draft.
delete_post write DELETE /posts/{post_id} Delete a draft, scheduled or published post, with its comments.

Publishing

Tool Access REST route What it does
publish_post write POST /posts/{post_id}/publish Publish a draft now.
schedule_post write PUT /posts/{post_id}/schedule Publish a draft at a set time, on the web and by email.
cancel_scheduled_post write DELETE /posts/{post_id}/schedule Stop a scheduled post from going out; it stays as a draft.
unpublish_post write DELETE /posts/{post_id}/publish Take a published post off the web and back to drafts.

Sections and post tags

Tool Access REST route What it does
list_sections read GET /sections Sections of your publication, the separate newsletters inside it.
list_post_tags read GET /post-tags Tags your publication uses to group posts.
create_post_tag write POST /post-tags New tag for grouping your posts.
tag_post write PUT /posts/{post_id}/tags/{tag_id} Add a tag to a post, or take it off.

Comments

Tool Access REST route What it does
list_comments read GET /posts/{post_id}/comments Comments on a post, with author, text and whether you replied.
reply_to_comment write POST /comments/{comment_id}/replies Reply as you under a comment on your post.
pin_comment write PUT /comments/{comment_id}/pin Pin a comment to the top of your post's comments, or unpin it.
like_comment write PUT /comments/{comment_id}/like Like or unlike a comment or a Note as you.
remove_comment write POST /comments/{comment_id}/remove Remove someone's comment from your post as the publication's moderator.
delete_comment write DELETE /comments/{comment_id} Delete one of your own comments, Notes or Note replies.

Notes

Tool Access REST route What it does
list_notes read GET /notes Notes posted by you or any other Substack user, newest first, with likes and replies.
list_note_replies read GET /notes/{note_id}/replies Replies under a Note, with author, text and likes.
publish_note write POST /notes Short post in Substack Notes, now or at a set time, optionally with a link card to one of your posts.
reply_to_note write POST /notes/{note_id}/replies Reply as you under any Note or Note reply.

Subscribers

Tool Access REST route What it does
list_subscribers read GET /subscribers Subscribers with plan, start date and activity rating.
add_subscriber write POST /subscribers Add one email to your list as a free subscriber.
remove_subscriber write DELETE /subscribers/{email} Remove one email from your subscriber list.
export_subscribers read GET /subscribers/export The full subscriber list as a CSV download link, with the row count.
list_subscriber_tags read GET /subscriber-tags Tags you use to group subscribers.
create_subscriber_tag write POST /subscriber-tags New private tag for grouping subscribers.
tag_subscribers write POST /subscriber-tags/{tag_id}/subscribers Put a tag on subscribers by email, or take it off.

Stats

Tool Access REST route What it does
get_publication_stats read GET /stats Total and paid subscribers, new subscribers, paid cancellations and estimated revenue for a period.
get_daily_stats read GET /stats/daily Day by day: total and paid subscribers, followers and views.
get_audience_insights read GET /audience Where your subscribers live, and which publications share your readers.
list_traffic_sources read GET /traffic-sources Where your publication's visitors came from in a period, most views first.
get_post_stats read GET /posts/{post_id}/stats Recipients, opens, open rate, views, likes, restacks and subscribers gained.
get_post_traffic read GET /posts/{post_id}/traffic Where a post's readers came from: sources, categories and devices.
get_post_engagement read GET /posts/{post_id}/engagement Who liked and commented on a post, and the subscribers it won and lost.

Recommendations

Tool Access REST route What it does
list_recommendations read GET /recommendations Publications you recommend, or that recommend you, with the subscribers each sent.
recommend_publication write PUT /recommendations/{publication_id} Recommend another publication to your readers, or stop recommending it.

Reader, feed and messages

Tool Access REST route What it does
get_feed read GET /feed Your Substack home feed: posts and Notes, For you or Following.
search_substack read GET /search Search all of Substack for people, posts and Notes.
list_activity read GET /activity Your Substack notifications: replies, likes, new followers, restacks, newest first.
get_unread_counts read GET /unread-counts How many unread direct messages, message requests and activity notifications you have.
list_conversations read GET /conversations Your direct-message conversations, latest first, with the last message.
get_conversation read GET /conversations/{conversation_id} The latest messages of one direct-message conversation, oldest first.
send_message write POST /messages Send a direct message on Substack, in a conversation or to a person.
follow_user write PUT /users/{user}/follow Follow or unfollow a Substack user, so their Notes reach your feed.
like_post write PUT /posts/{post_id}/like Like or unlike any Substack post as you.
list_subscriptions read GET /subscriptions Publications you subscribe to, with your plan and whether their emails reach you.

Never built, on purpose: your Substack sign-in, email, profile and account settings, your publication's address or ownership, and anything that moves money (paid plans, payouts, comps). Every request passes a guard that refuses writes to those Substack paths.

Configuration

Variable Default What it does
SUBSTACK_MCP_HOME ~/.config/substack-api-mcp Folder of session.json.
SUBSTACK_AUTH_TOKEN Substack substack.sid cookie, instead of substack-mcp connect.
SUBSTACK_PROXY Proxy for every request to Substack, e.g. http://user:pass@host:port.
SUBSTACK_MCP_READ_ONLY 1 hides and refuses every tool that changes Substack (same as --read-only).
SUBSTACK_MCP_TOKEN Bearer token required by serve; needed to listen beyond 127.0.0.1.
SUBSTACK_MCP_HOST, SUBSTACK_MCP_PORT 127.0.0.1, 8787 Address of serve.

Security and your Substack account

  • No password anywhere. Substack signs in by an emailed code. Only cookies are read, and only those of substack.com.
  • The session file is yours. session.json is created with mode 0600. Anyone who has it can act as you on Substack, so keep it out of git and shared folders.
  • Local by default. serve binds to 127.0.0.1 and refuses another address without SUBSTACK_MCP_TOKEN.
  • substack.com only. Requests and redirects outside substack.com and its subdomains are refused.
  • Read-only mode. --read-only or SUBSTACK_MCP_READ_ONLY=1 exposes the 25 read tools only.
  • Actions reach real people. Published and scheduled posts are emailed to your subscribers, and Notes, comments, replies and DMs go out under your name. MCP clients ask before running them; a script does not, so try it on a test publication first.
  • Add only people who asked. add_subscriber sends Substack's welcome email. Importing people who did not sign up breaks Substack's rules.

Errors

The REST API answers {"error": "<code>", "message": "..."}. The MCP server returns the same code and message as a tool error.

Status Code What to do
401 auth_required, auth_expired No session, Substack signed you out, or it asks you to sign in again before a sensitive action (subscriber lists). Sign out and in at substack.com in Chrome, then run substack-mcp connect.
403 app_permission_denied Your role in the publication cannot do this.
403 scope_denied The server runs read-only, or the call would change your account settings.
403 target_blocked Substack refused the request or showed a bot check. Open substack.com in Chrome once, then retry.
409 captcha_required Substack asks for a CAPTCHA. Solve it in Chrome.
422 invalid_input An argument is missing or wrong; the message names it.
429 rate_limited Substack is limiting the account. Wait a few minutes.
502 target_error Substack answered with an error, or changed its site. git pull for the latest tools, or open an issue.

Self-hosted or hosted

This repo Cotlet, hosted
Tools the same 50 the same 50
Runs on your computer, while it is on in the cloud, 24/7
Substack session your Chrome, cookies in a local file your own Cloud Browser in the EU, profile encrypted with AES-256-GCM
When Substack changes its site git pull once we have fixed it fixed for you, every tool tested daily
Approvals your MCP client asks per call posts, Notes, replies and DMs wait for your OK, switchable per tool
Works with MCP clients on your machine, local scripts Claude.ai and ChatGPT connectors, n8n, Make, Zapier, any HTTP client
Who may use it you, for personal use teams, businesses, agencies
Price free free 7-day trial with 2,000 credits, no card; then one-time packs from $29 for 5,000 credits

On the hosted version a successful call uses 20 to 30 credits (25 for most tools), failed calls are free, and the Cloud Browser uses 50 credits a day while it has an app enabled for agents. Full pricing.

FAQ

Does Substack have an official API? Only a limited one (checked 11 Oct 2026). Its Developer API looks up public creator profiles after approval, and there is no public API to write posts or Notes, manage subscribers or send email. RSS reads public posts only. Details and sources: cotlet.dev/integrations/substack-api-mcp-server.

Is there a Substack MCP server for Claude, ChatGPT or Cursor? Yes, this one runs on your machine: substack-mcp mcp for Claude Code, Claude Desktop, Cursor and Codex. ChatGPT and Claude.ai connectors need a public HTTPS URL, which the hosted Substack MCP server gives you.

Can I schedule Substack posts from Markdown? Yes: python examples/draft_from_markdown.py essay.md --publish-at 2026-11-03T08:00:00-05:00. The Substack editor does not import Markdown. The script sends the file as a draft with its headings, bold text, links, lists and code blocks, then schedules it. Images have to be public URLs.

Can it schedule Notes as well as posts? Yes. publish_note takes a publish_at time and the Note waits in your scheduled Notes until then. A link card needs a published post.

Can I export my Substack subscribers to CSV? Yes: python examples/export_subscribers.py --out subscribers.csv gives the same CSV as Export in the subscriber dashboard. list_subscribers.py --plan paid or --since 2026-10-01 gives a filtered list with plan, join date and activity rating.

Can I export my posts as Markdown? Yes, one at a time with examples/get_post.py --post-id <id> --out post.md, including the paywalled part and drafts of your own publication. list_posts.py gives the IDs.

Is this a Substack scraper? It does what people want a scraper for (export posts, subscribers and stats) but does not parse HTML in a headless browser. It calls the same JSON endpoints the Substack web app calls, in your own account and publication.

Why does it need my Chrome? To read your Substack session. Substack signs in by an emailed code, so there is no password to give it anyway. Chrome is used once per sign-in, not for each call.

What happens when Substack signs me out? Calls fail with auth_expired. With Chrome remote debugging on, the server reads the new cookies on the next call (click Allow when Chrome asks); otherwise run substack-mcp connect again. Subscriber lists sometimes ask for a fresh sign-in even with a valid session: sign out and in again at substack.com, then connect.

Does it need a paid Substack plan? No. Revenue, paid subscribers and cancellations exist only for publications with paid subscriptions.

Can it email all my subscribers? Only by publishing or scheduling a post, the way the editor sends it. Substack emails a post once; edits after the send show on the web and in the app.

Can it run on a VPS or in Docker? Technically yes, with SUBSTACK_AUTH_TOKEN. Substack then sees your session from a datacenter IP, which looks less like you, and the license allows it only on a server you alone use. For an always-on setup the hosted version keeps your session in your own Cloud Browser on a fixed IP.

Will my Substack account get banned? Requests carry Chrome's fingerprint and go out one at a time, as you would click. Nobody can promise a platform never acts. Keep to your normal use and do not add or message people in bulk.

Can I use it at work or for my clients? No, the license is for personal use by one person. For a team, a company or client work, use Cotlet or write to contact@cotlet.dev.

Is it open source? The source is public, the license is not an open source license: personal use only, no business use, no hosting for others.

A tool broke or I need another one. Open an issue with the tool name and the error message.

License

Cotlet Personal Use License 1.0. In short:

  • Allowed: one person, on their own computers, with their own Substack accounts, for their own purposes. Changing the code for yourself, and a GitHub fork under the same license.
  • Not allowed: any use in or for a company or organization, your own business included; letting other people use your instance (shared server, team machine, cloud service); selling or offering it or a service built on it.

Earlier versions of this repo (the scripts for the hosted API, up to commit 4399e88) stay MIT.

Cotlet

Maintained by Cotlet (@cotletdev). Substack is a trademark of its owner. This project is unofficial and not affiliated with, endorsed or sponsored by Substack.