squad-mobile
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from multiple external sources, which presents a surface for potential indirect prompt injection.\n
- Ingestion points: The agent is instructed to process input from deep links, push payloads, and backend API contracts, as well as descriptive user input for feature implementation (located in
SKILL.mdandreferences/design-platform-and-lifecycle-gates.md).\n - Boundary markers: The skill lacks formal boundary delimiters for untrusted data in its prompt instructions, though it does categorize these sources as untrusted.\n
- Capability inventory: The skill allows for the execution of build commands, process management of simulators and build daemons, and file system modifications for app logic implementation.\n
- Sanitization: Positive security instructions are present, requiring the agent to validate payloads and redact sensitive information from logs and crash reports as defined in
references/mobile-security-performance-testing-and-release.md.
Audit Metadata