helios-studio
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions direct the agent to execute
npx @helios-project/cli studio. This command downloads and executes the Helios command-line interface from the npm registry to start a local development server. - [INDIRECT_PROMPT_INJECTION]: The skill facilitates agent interaction with composition schemas and input props through a Model Context Protocol (MCP) server, which presents a potential surface for indirect injection if external composition files contain malicious instructions.
- Ingestion points: The agent ingests data from local
HeliosSchemafiles and input props through the studio's MCP interface. - Boundary markers: The skill does not define specific delimiters or instructions to prevent the agent from obeying instructions embedded within the composition data.
- Capability inventory: The associated MCP server allows the agent to inspect schemas, update props, trigger renders, and create new files in the local project directory.
- Sanitization: There are no explicit mentions of sanitization or validation of the input props or schemas before they are processed by the agent.
Audit Metadata