xbird

Use when the user asks to tweet, post threads, read tweets, search Twitter/X, check mentions, manage engagement (like/retweet/bookmark), update profile (bio, avatar, banner), upload media, or interact with Twitter accounts. Triggers: twitter, tweet, post, thread, timeline, mentions, followers, following, likes, retweet, bookmark, profile picture, bio.

Do not paste your Twitter session cookies or your wallet private key into a third-party skill unless you fully trust and can verify the code and publisher. The SKILL.md asks you to run an unpinned npm package via npx and to store sensitive tokens in your settings; that package will execute arbitrary code locally. Before installing: (1) verify the package source and repository (read its code, release tags, and who publishes it), (2) prefer official OAuth/API keys rather than raw session cookies, (3) never give out your wallet private key — use an intermediate payment/account that you control with limited funds or a signing-only flow, and (4) if you must test, run any fetched code in an isolated environment (VM/container) and avoid adding it to your main agent config. Given the metadata/instruction mismatches and sensitive requests, proceed only after independent code review or prefer an official, documented integration.

Type: OpenClaw Skill Name: xbird Version: 0.1.1 The skill is classified as suspicious due to two main reasons. Firstly, it instructs the user to provide highly sensitive credentials, including Twitter session tokens (`XBIRD_AUTH_TOKEN`, `XBIRD_CT0`) and a crypto wallet private key (`XBIRD_PRIVATE_KEY`), to an external `npx` package (`@checkra1n/xbird`). This introduces a significant supply chain risk and exposes the user to potential credential theft if the external package is compromised. Secondly, the `update_profile_image`, `update_profile_banner`, and `upload_media` tools in `SKILL.md` accept absolute file paths, creating a local file access vulnerability where an agent could be prompted to read arbitrary local files from the filesystem.