xiaohongshu-post
llms.txt1 starsChanged 21 days agoMITView on GitHub
An agent skill for publishing image-and-text notes to Xiaohongshu (小红书 / RED) from a logged-in Chrome profile via Playwright over CDP. For AI agents: load this when the user asks to post/publish a note to 小红书 / RED, or wants a public share link for a note they published.
In this repositoryOne other file it gives its agents.
- xiaohongshu-postSkill
The file
We split this llms.txt into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 26
Do NOT use for: video notes, live, comment automation, multi-account matrix publishing.
≈ “Common” is an estimate: at least 1 in 5 of the 3,592 llms.txt files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
# xiaohongshu-post
An agent skill for publishing image-and-text notes to Xiaohongshu (小红书 / RED) from a
logged-in Chrome profile via Playwright over CDP.
For AI agents: load this when the user asks to post/publish a note to 小红书 / RED, or wants a
public share link for a note they published.
## What it gives you
- A browser bootstrap that keeps the **login** across agent turn boundaries. The browser
*process* is reaped no matter how you start it (measured twice) — what persists is the
**profile directory** (`$PROFILE_DIR/Default/Cookies`, holds `id_token`). Re-launch Chrome
against the same profile dir each turn and it comes back logged in. Never use a fresh
profile per turn.
- `scripts/publish_note.js`: image upload, title/body filling, insertion of REAL topic tags via
the Tiptap suggestion dropdown, draft screenshots. Does not publish unless `--publish` is passed.
- `scripts/verify_note.js`: extracts note id / `xsec_token` / short link and proves public
visibility from a fresh logged-out context (curl cannot — the page is client-rendered).
- `references/selectors.md`: every selector, API field, and pitfall with its root cause.
- `references/copy-guide.md`: title/body/image specs (title ≤20 chars, body ≤1000, cover 3:4).
## When to activate (triggers)
- "发小红书" / "上传小红书" / "帮我发个笔记" / post to Xiaohongshu / RED.
- Automating a publish where the account is already logged in.
- Recovering a working public link for a published note.
Do NOT use for: video notes, live, comment automation, multi-account matrix publishing.
## Files (read these)
- SKILL.md — the workflow, the three mandatory pre-checks, and the pitfall quick-reference.
- references/selectors.md — selectors, API endpoints, field meanings, verified platform limits.
- references/copy-guide.md — copy template + image spec + cover layout.
- scripts/bootstrap_browser.sh — start Chrome with CDP + attach playwright-cli.
- scripts/publish_note.js — fill the draft (add `--publish` to click 发布).
- scripts/verify_note.js — extract links + verify public visibility.
## Run it
bash scripts/bootstrap_browser.sh .tmp/xhs-cdp 9222
node scripts/publish_note.js --title "..." --body-file body.txt --images a.png,b.png --topics "X,Y"
node scripts/verify_note.js --expect "标题前几个字"
Exit 0 = ok, 1 = error, 2 = bad input, 3 = published but NOT publicly visible.
## Install
npx skills add CatsMansion/xiaohongshu-post
Repo: https://github.com/CatsMansion/xiaohongshu-post
## License
MIT.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

