ink
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for scaffolding new terminal applications using npx assistant-ui@latest create --ink.
- [EXTERNAL_DOWNLOADS]: The documentation recommends installing several Node.js packages from the @assistant-ui namespace along with ink and react.
- [INDIRECT_PROMPT_INJECTION]: The skill builds a chat interface which inherently processes untrusted data from both users and AI models.
- Ingestion points: Untrusted data is ingested and rendered via ThreadPrimitive.Messages and MessagePrimitive.Parts as documented in SKILL.md and references/primitives.md.
- Boundary markers: The skill uses MarkdownText for rendering but does not specify explicit boundary markers or instructions to ignore embedded commands in ingested data.
- Capability inventory: The skill supports writing to the local file system through createFileStorageAdapter (references/adapters.md) and making network requests via AssistantChatTransport (references/custom-backend.md).
- Sanitization: No explicit sanitization or content filtering for external data is documented for the provided primitives.
Audit Metadata