| name | add-post |
| description | Add a new blog post or article to the documentation collection. Use when user provides a URL to a blog post, article, or tweet linking to one. |
| argument-hint | ["url"] |
Add new blog post
Add a new blog post or article to the documentation collection.
Input
- URL to a blog post/article or to a tweet/discussion about a blog post
Steps
-
Check browser is activated:
- Call
tabs_context_mcp to verify browser connection
- If it fails, ask the user to enable browser:
- VS Code: Ensure Chrome extension is connected
- CLI: Run
claude --chrome or type /chrome
-
Find the canonical blog post page: If the input URL is:
- A tweet (x.com or twitter.com): Use browser to navigate to the tweet, extract the blog post link
- A direct blog post URL: Use that URL
- Medium.com or Substack.com: Use browser (WebFetch is blocked)
- Other: Use the provided URL
- For X.com (Twitter) links, it can be either a X post or X article - handle articles as posts
Important: For Twitter/X, Medium, Substack, and sites requiring JavaScript, use browser automation instead of WebFetch.
-
Extract blog post information:
- Title: The full blog post title
- Author/Source: The author name or blog/publication name
- Description: A concise summary (1-2 paragraphs) describing what the post covers
For pages requiring browser: Use navigate to open the page, then use javascript_tool to extract content. See CLAUDE.md > JavaScript DOM Extraction Patterns for selectors and examples.
If CAPTCHA appears, ask the user to complete it manually.
-
Determine category: Based on the extracted content, automatically determine the category:
- Trading/Finance (goes to
source/learn/blog-posts.rst) - for posts about algorithmic trading, quantitative finance, market analysis, portfolio management, risk management, backtesting, etc.
- AI/ML (goes to
source/learn/ai-and-machine-learning.rst) - for posts primarily about machine learning, deep learning, neural networks, or AI techniques applied to trading
Not trading related: Don't add this post. If the user specifically asked to add this post, ask user for confrimation.
Only ask the user if the content is ambiguous (e.g., equally about ML techniques AND trading strategies). If clearly one category, proceed without asking.
-
Add to the appropriate .rst file: Use this exact format (matching existing entries):
Title of the Blog Post
----------------------
Description paragraph(s) here. Keep it informative but concise.
Two paragraph summary of the post here.
`Read the blog post <https://url-here>`__.
Or with author attribution:
Title of the Blog Post
----------------------
Description paragraph(s) here.
Two paragraph summary of the post here.
By Author Name.
`Read the blog post <https://url-here>`__.
Important formatting notes:
- The underline of dashes must be at least as long as the title
- There should be a blank line after the title underline
- There should be a blank line before the "Read the blog post" link
- The link format uses double underscores at the end:
__
- Use "Read the blog post" or "Read the article" or "Read more" as appropriate
- Add the new entry at the END of the file
If the source of the link is a discussion like a tweet, then include a paragraph with a link to that tweet with the comment "Mentioned by XXX in this discussion" and include what they say about it.
-
Save as PDF: Follow the procedure in README-browser.md > Saving PDFs > How it works using:
- URL: the blog post URL
- FILENAME:
<slugified-title>.pdf
- Save to
articles/ directory
-
Update the article index: After adding the entry to the .rst file and saving the PDF, update articles/index.json. See .claude/docs/paper-index.md for the full schema. Append a new entry:
{
"title": "Blog Post Title",
"authors": "Author Name or Publication",
"publication_year": 2026,
"download_page": "https://canonical-url",
"source_file": "target-file.rst",
"included_in_index_at": "YYYY-MM-DD",
"filename": "slugified-title.pdf",
"downloaded_at": "YYYY-MM-DD",
"download_failure_reason": null,
"short_description": "One paragraph (2-3 sentences) describing what the post covers and its key takeaway."
}
Use the same source_file you added the entry to (e.g. the topic file such as volatility.rst), not a literal blog-posts.rst. If the PDF could not be saved, set filename and downloaded_at to null and set download_failure_reason to the appropriate reason (requires_browser, paywall, not_attempted).
Keep the array sorted by source_file then title.
-
Commit and push:
- Stage the modified
.rst file, the PDF (force-add with git add -f, since articles/*.pdf is gitignored), and articles/index.json
- Commit with message: "Add: {Blog Post Title}"
- Push to master branch
