service-de-waba-integrate
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
echocommand to write a JSON result to a temporary file (/tmp/linking-waba-result.json). The content of this file is determined by a static mapping of user responses (Yes/No/Skip) and does not involve the interpolation of untrusted external input into the shell command execution. - [EXTERNAL_DOWNLOADS]: The skill provides instructions directing the user to Meta's official domain (
business.facebook.com). This is a well-known service and the legitimate platform for managing WhatsApp Business Account partnerships. - [INDIRECT_PROMPT_INJECTION]: The skill ingests
WABA_IDandSF_BM_IDas inputs from the calling process to render personalized instructions. - Ingestion points: Variables
{WABA_ID}and{SF_BM_ID}inSKILL.md. - Boundary markers: Instructions are rendered as static markdown text directly to the user.
- Capability inventory: The skill has the ability to write a status file to the local filesystem.
- Sanitization: While no explicit sanitization is performed on these variables, they are only used for text rendering in instructions and are not passed to the shell command in Stage 3, minimizing the risk of prompt or command injection.
Audit Metadata