kapso-whatsapp
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes inbound WhatsApp messages, which are untrusted data from external sources.
- Ingestion points: Inbound webhook payloads, specifically the message.text.body field and media content (SKILL.md).
- Boundary markers: None specified in the instructions for handling message content.
- Capability inventory: The skill allows for sending messages, downloading media, and interacting with a user registry (SKILL.md).
- Sanitization: The provided recipes do not include explicit steps for sanitizing or validating external input before processing.
- [EXTERNAL_DOWNLOADS]: The documentation includes instructions to clone an external repository from GitHub and install its dependencies to provide fallback functionality (SKILL.md).
- Evidence: git clone --depth=1 https://github.com/gokapso/agent-skills /tmp/kapso-skills followed by npm i.
Audit Metadata