The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/linkedin-post.js executes execSync to run the clawdbot command-line tool, which is used to manage the browser lifecycle.
[REMOTE_CODE_EXECUTION]: The script uses Chrome DevTools Protocol (CDP) Runtime.evaluate to execute arbitrary JavaScript within the browser context. This allows the script to manipulate the LinkedIn page and inject user-supplied content from the --text argument into the DOM.
[CREDENTIALS_UNSAFE]: The skill instructions specify a dependency on a session cookie file at cookies/linkedin-zeusbycreatoros.json. Accessing or relying on persistent session cookies in specific file paths is a risk factor for credential exposure.
[PROMPT_INJECTION]: The skill possesses a significant attack surface for indirect prompt injection.
Ingestion points: The --text command-line argument accepts untrusted data from the agent's context.
Boundary markers: No clear boundary markers or instructions to ignore embedded commands are present in the automation script.
Capability inventory: The skill can perform shell command execution (execSync), browser navigation, and arbitrary JavaScript execution in a logged-in session.
Sanitization: While the script performs basic HTML escaping for the LinkedIn editor, the dynamic evaluation of user-controlled text in the browser environment remains a risk.

linkedin-post