The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses bun run as the primary interface for its operations. This is standard behavior for a Node-based tool using Bun and is used to execute the provided scripts/post.ts script locally.
[DATA_EXFILTRATION]: The script performs authenticated HTTP POST requests to official social media endpoints (api.x.com and api.linkedin.com) to publish content. This is the primary and stated purpose of the skill. To prevent accidental data leaks or unauthorized posting, the skill implements a 'dry run' mechanism by default. It only performs network calls when the --confirm flag is explicitly provided. Sensitive environment variables (X_BEARER_TOKEN and LINKEDIN_ACCESS_TOKEN) are read into memory but are masked using the maskAuth function before being displayed in dry-run logs.
[SAFE]: The skill demonstrates security best practices for social media integration by enforcing human-in-the-loop confirmation before destructive or public actions, using stateless environment variables for credentials rather than hardcoded secrets, and minimizing external dependencies by using built-in platform features like fetch.

social-poster