swarm: Command for Claude Code

/add_post

Scaffold and publish an original blog post to swarm-ai.org/blog/ — handles MkDocs slug generation, metadata headers, nav wiring, and financial-disclaimer enforcement. Use for manually-written essays and research notes; use /eval_writeup instead to auto-generate a post from a Prime Intellect eval run.

Usage

/add_post <title> [source-file]

Examples:

• /add_post "Collusion Detection Deep Dive" (write from scratch)
• /add_post "When Agents Collapse" path/to/draft.md (import existing file)

Behavior

1. Derive slug from the title:

• Lowercase, replace spaces with hyphens, strip punctuation.
• e.g. "Collusion Detection Deep Dive" → collusion-detection-deep-dive

1. Create the post file at docs/blog/<slug>.md:

• If <source-file> is provided, copy its content and fix image paths to be relative to docs/blog/ (e.g. ../papers/figures/...).
• If no source file, create a skeleton with the title as H1, a *subtitle* italic line, a --- separator, and placeholder sections.

1. Update the blog index at docs/blog/index.md:

• Add a bullet entry linking to the new post with the title and a one-line description.
• Insert newest post at the top of the list.

1. Update mkdocs.yml nav:

• Add the new post under the Blog: nav section.
• Place it after the index entry.

1. Verify the build:

• Run python -m mkdocs build --strict and report any errors.
• If the build fails (e.g. broken image path), fix the issue before proceeding.

1. Show what changed:

• Print the list of files created/modified.
• Do NOT commit or push — let the user decide when to ship (via /ship or manual commit).

Constraints

• Never overwrite an existing post file without asking.
• Always verify image paths resolve correctly (relative to docs/blog/).
• If docs/blog/index.md doesn't exist, create it with a header.
• If the Blog: section doesn't exist in mkdocs.yml nav, create it.