configure-formatting

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill processes project configuration files and external web content, creating a surface for indirect prompt injection.
  • Ingestion points: Reads local configuration files such as biome.json, package.json, pyproject.toml, and Cargo.toml. It also fetches version data from external URLs including biomejs.dev, prettier.io, astral.sh, and releases.rs.
  • Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present when processing the content of ingested files or web data.
  • Capability inventory: The skill can execute shell commands via Bash (e.g., to install packages or run formatters) and modify project files using Write and Edit tools.
  • Sanitization: There is no evidence of input validation or sanitization for fetched web content or local configuration data before it is processed by the agent.
  • [COMMAND_EXECUTION]: Executes local commands for project discovery, package management, and formatter operations.
  • Uses Bash to run discovery commands like find and grep to identify project types and existing configuration files.
  • Executes package managers (npm, bun, uv) to install developer tools and dependencies.
  • Runs formatter-specific commands such as biome format, ruff format, and cargo fmt to check compliance and apply fixes.
  • [EXTERNAL_DOWNLOADS]: Fetches version information and installs packages from trusted and well-known sources.
  • Downloads configuration and version metadata from official project sites for Biome, Prettier, Ruff, and Rust.
  • Installs standard formatting packages from NPM (@biomejs/biome, prettier) and PyPI (ruff, black).
  • References pre-commit hook repositories from official GitHub organizations associated with the formatting tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 09:33 AM