apple-contacts
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes
osascriptto execute AppleScript commands, allowing the agent to interact directly with the macOS system and the Contacts application. - [INDIRECT_PROMPT_INJECTION]: The skill retrieves data from an external source (the user's address book) which is then processed by the agent. This represents an indirect prompt injection surface where malicious contact information could influence agent behavior.
- Ingestion points: Data is ingested via shell command output from
osascriptqueries targeting theContactsapplication, as described inSKILL.md. - Boundary markers: None identified. Data retrieved from contacts is not delimited or wrapped in instructions to ignore embedded commands.
- Capability inventory: The skill has the capability to execute shell commands (
osascript) on the host system. - Sanitization: No sanitization or escaping of contact data (names, phone numbers, emails) is performed before the information is returned to the agent context.
Audit Metadata