bird-fast

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt documents and specifically supports passing Twitter auth cookies/tokens via CLI flags (--auth-token, --ct0) and shows/env-var export examples, which encourages embedding secret values verbatim in generated commands or code (even though browser-cookie usage is an alternative), so an LLM following the skill could be required to output secrets directly.

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly fetches and ingests user-generated content from the open web (X/Twitter) as shown in SKILL.md via commands like "bird read", "bird replies", "bird thread", "bird search", and "bird mentions", and those results are used in automation workflows (e.g., replies, posting, monitoring) that could drive subsequent actions.