skills/microsoft/aspire-skills/aspire/Gen Agent Trust Hub

aspire

Pass

Audited by Gen Agent Trust Hub on Jul 31, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [Remote Code Execution]: The skill documents the installation of the Aspire CLI using curl -sSL https://aspire.dev/install.sh | bash. While piping remote scripts to a shell is a common security consideration, this specific script originates from the official domain for the Aspire technology and is the standard installation method provided by the vendor.
  • [External Reference]: The skill links to official resources on dotnet.microsoft.com and aspire.dev for software prerequisites and documentation. These are well-known, official domains for the tools described.
  • [Indirect Prompt Injection Surface]: The skill ingests data from local workspace files, such as .csproj and aspire.config.json, to determine routing and project state.
  • Ingestion points: Reads project configuration and source files (SKILL.md).
  • Boundary markers: The skill includes instructions to enforce safety guardrails and use --non-interactive flags for agent execution (SKILL.md).
  • Capability inventory: Executes shell commands via the Aspire CLI for project lifecycle management, including start, deploy, and add (SKILL.md).
  • Sanitization: Employs non-interactive execution and encourages the use of project-local skills for scenario-based guidance.
  • [Command Execution]: The skill instructs the agent to use the Aspire CLI to manage distributed applications. It enforces the use of specific flags like --non-interactive to maintain consistent behavior during automated tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 31, 2026, 05:03 AM
Security Audit — agent-trust-hub — aspire