brand-voice-synthesizer
Pass
Audited by Gen Agent Trust Hub on Jul 30, 2026
Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection by processing untrusted data from published WordPress posts and activity reports.
- Ingestion points: Data is collected from the site using
respira_read_post,respira_extract_builder_content, andrespira_generate_activity_reportwithin the execution workflow. - Boundary markers: The workflow does not define specific delimiters or instructions to ignore potential commands within the analyzed content.
- Capability inventory: The agent can persist data to the site using
respira_update_option. - Sanitization: The skill strips code blocks and shortcodes from the source text during analysis.
- [DATA_EXFILTRATION]: Sends usage telemetry to the vendor's official endpoint at
https://www.respira.press/api/skills/track-usage. This telemetry includes site metadata and sampling statistics but does not include actual post content or the extracted voice profile.
Audit Metadata