Skills
skills/aradotso/claude-code-skills/claude-code-config-trailofbits/Gen Agent Trust Hub

claude-code-config-trailofbits

Fail

Audited by Gen Agent Trust Hub on May 17, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill instructs the user to download and execute scripts directly from the internet using pipes to shell environments.
  • Pipes the Rust toolchain installer from https://sh.rustup.rs to sh.
  • Pipes the LM Studio installer from https://lmstudio.ai/install.sh to bash.
  • [COMMAND_EXECUTION]: The skill modifies local shell configuration files to establish persistence for custom commands and aliases.
  • Appends a new alias to ~/.zshrc using an echo command: alias claude-yolo="...".
  • [PROMPT_INJECTION]: The skill advocates for the bypass of the agent's built-in security and permission systems.
  • Recommends the use of the --dangerously-skip-permissions flag in a shell alias to disable the standard interactive permission prompts normally required for tool execution.
  • [EXTERNAL_DOWNLOADS]: The skill downloads configuration and additional executable logic from external repositories.
  • Fetches the primary configuration suite from https://github.com/trailofbits/claude-code-config.git.
  • Downloads external skill collections from https://github.com/trailofbits/skills.git and https://github.com/trailofbits/skills-curated.git into the agent's persistent skill directory.
Recommendations
  • HIGH: Downloads and executes remote code from: https://sh.rustup.rs, https://lmstudio.ai/install.sh - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
May 17, 2026, 01:34 PM
Security Audit — agent-trust-hub — claude-code-config-trailofbits