skills/dashed/claude-marketplace/ty/Gen Agent Trust Hub

ty

Pass

Audited by Gen Agent Trust Hub on Aug 18, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill documentation includes a standalone installation command (curl -LsSf https://astral.sh/ty/install.sh | sh) that fetches and executes a shell script from Astral's official domain. Astral is a well-known provider of Python developer tooling, and this installation method is a standard distribution pattern for their services.
  • [COMMAND_EXECUTION]: The instructions direct the agent to run the ty binary and its subcommands (e.g., ty check, ty server, ty explain) to perform type checking and environment inspection. These operations are essential to the primary purpose of the skill and utilize standard command-line interfaces.
  • [INDIRECT_PROMPT_INJECTION]: The skill enables the agent to process and analyze content from local source code and project configuration files. 1. Ingestion points: Python source files and project-specific configuration files (ty.toml, pyproject.toml) identified during ty check operations. 2. Boundary markers: No specific delimiters or instructions are used to isolate potentially untrusted content within these files from the agent's logic. 3. Capability inventory: The agent can execute the ty CLI tool, view rule explanations, and interact with a language server. 4. Sanitization: The skill relies on standard tool-based parsing without additional input validation layers.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 18, 2026, 12:24 PM
Security Audit — agent-trust-hub — ty