boha-cli

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md and references/collections.md explicitly show boha opening and returning public asset URLs (e.g., "boha show zden/level_4 --open" and p.asset_url() -> GitHub raw URL, plus links like privatekeys.pw), which means the skill fetches and displays untrusted third‑party web content (images/pages) that the agent is expected to read and which could influence subsequent actions.