unifonic
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user data (message bodies and recipient phone numbers) which is then interpolated into shell-executed
curlcommands. This creates a surface where adversarial input could attempt to influence the agent's behavior or manipulate the shell command structure. - Ingestion points: The
RecipientandBodyparameters in thecurltemplates withinSKILL.mdare populated by user-provided data. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the command templates.
- Capability inventory: The skill enables network operations via shell execution of
curl. - Sanitization: The skill does not define specific sanitization or validation logic for the input, relying on the underlying agent platform's safety guardrails.
Audit Metadata