evolution-api
Pass
Audited by Gen Agent Trust Hub on Jun 27, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists entirely of instructional markdown and API reference documentation. No executable scripts, automated tools, or hidden payloads are included.
- [SAFE]: Hardcoded credential examples found in the documentation (e.g., in environment-variables.md) are clearly marked as placeholders or examples for self-hosting configuration and do not represent leaked secrets.
- [PROMPT_INJECTION]: The skill documents an indirect prompt injection surface inherent to its purpose as a messaging gateway. It enables the ingestion of untrusted external content (WhatsApp messages via webhooks) and provides capabilities to route this data to AI integrations such as OpenAI or Dify. (Ingestion points: references/events.md; Capability inventory: SKILL.md; Boundary markers/Sanitization: Not specified in documentation).
Audit Metadata