bird

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt documents and encourages passing cookie auth tokens directly via flags like --auth-token / --ct0 (command-line args) which would require an agent to include secret cookie values verbatim in generated commands, creating an exfiltration risk.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 1.00). The SKILL.md shows the CLI reads and searches public X/Twitter content (e.g., "Reading Tweets" commands like bird read <url-or-id>, timelines, bird search) which are untrusted, user-generated third-party sources and the skill also exposes engagement commands (e.g., bird reply, bird tweet, bird follow) that could be driven by or act upon that content, enabling indirect prompt injection.