bird

name: bird description: X/Twitter CLI for reading, searching, and posting via bird. Use when the user wants to read tweets, search X, inspect threads, mentions, timelines, bookmarks, or post/reply on X/Twitter. homepage: https://github.com/longlannet/bird metadata: { "openclaw": { "emoji": "🐦", "requires": { "bins": ["bird"], "config": ["config.env"], }, "install": [ { "id": "node-bird", "kind": "node", "package": "@steipete/bird", "bins": ["bird"], "label": "Install bird (node)", }, ], },

}

Use this skill for X/Twitter tasks through the local bird CLI.

When to use

Use this skill when the user wants to:

read a tweet or thread
search X/Twitter
inspect a user's tweets
check mentions, bookmarks, likes, followers, or following
post or reply on X

Quick start

bash scripts/install.sh
source config.env && bird read <tweet-id-or-url> --json
source config.env && bird search "openclaw" --json
source config.env && bird mentions

Core bird installation is upstream at https://bird.fast/. This skill uses a local wrapper install path around the bird CLI for OpenClaw workflows.

Workflow

Confirm authentication is working with bird check or bird whoami.
Run the smallest useful read/search command first.
Use --json or --plain when stable output matters.
Post or reply only when the user explicitly wants a write action.

Notes

Core bird installation is upstream at https://bird.fast/; if upstream setup changes, follow upstream.
scripts/install.sh installs or resolves the local bird CLI and runs a smoke test, but it does not install Node.js/npm or create fresh auth cookies for you.
config.env is runtime config and should not be committed.
When migrating to another machine, reinstall bird there and copy config.env only between machines you control.
Run from the skill root so config.env is loaded correctly.
Prefer --json for parsing and summarization.
Re-run scripts/install.sh if bird is missing.