tailrocks-swift-rust-core-boundary
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides comprehensive architectural guidance focusing on separation of concerns and data integrity between Swift and Rust. It includes defensive references like
runtime-trust.mdwhich explicitly forbids untrusted data from altering the agent's scope or exfiltrating credentials. - [INDIRECT_PROMPT_INJECTION]: The skill guides the agent in processing untrusted application code and architectural requests.
- Ingestion points: Code revisions, paths, and target platforms defined in
SKILL.md. - Boundary markers: Guided by
runtime-trust.md, which establishes that repository content is untrusted and cannot override security rules. - Capability inventory: The agent performs architecture review and code implementation, which may involve tool usage for analysis.
- Sanitization: Mandates the scrubbing of secrets and citation of locations rather than reproduction of sensitive data.
Audit Metadata