monskill
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill mentions official CLI tools for well-known services, specifically
@alchemy/cli,@getpara/cli, andenvio-cloud. These tools are used for blockchain infrastructure, MPC wallet management, and data indexing. - [COMMAND_EXECUTION]: References shell commands for interacting with external services, such as
alchemy auth,para login, andenvio-cloud login. The skill explicitly implements a safety best practice by instructing the agent never to run installation or authentication commands on the user's behalf, requiring the agent to surface a prompt and wait for user action. - [INDIRECT_PROMPT_INJECTION]: As a router skill, it selects further instructions based on user-provided task descriptions, creating a surface for indirect prompt injection.
- Ingestion points: User-provided task descriptions in the prompt determine which sub-skills (e.g.,
scaffold/,wallet/) are fetched by the agent. - Boundary markers: None identified in this top-level router file.
- Capability inventory: The referenced sub-skills include capabilities for package installation, network-based CLI operations, and blockchain transactions.
- Sanitization: Not explicitly mentioned in this router file.
Audit Metadata