hugging-face-trackio

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

• Third-party content exposure detected (high risk: 0.75). The required runtime workflow for the agent (“poll for alerts” / “inspect metrics”) uses trackio list alerts ... --space <space_id_or_url> / trackio get ... --space / trackio query ... --space, which fetches outsider-authored free text from a remote HF Space (e.g., alert title/text) and injects it into the agent’s LLM context via the CLI/JSON output.