Skills
Navigate
skills/code-yeongyu/oh-my-openagent/lsp-setup/Gen Agent Trust Hub

lsp-setup

Fail

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: Instructions in SKILL.md and references/julia/README.md direct the installation of prerequisites through piped shell execution from remote URLs.
  • Evidence: curl -fsSL https://bun.sh/install | bash in SKILL.md.
  • Evidence: curl -fsSL https://install.julialang.org | sh in references/julia/README.md.
  • [DYNAMIC_EXECUTION]: The verification script uses runtime path resolution to dynamically load modules.
  • Evidence: scripts/verify-lsp.ts uses import() on paths resolved via a findUp search for packages/lsp-tools-mcp/src/tools.ts and other engine files.
  • [COMMAND_EXECUTION]: The skill manages environment setup through various system and language-specific package managers.
  • Evidence: Commands for npm, pip, brew, dotnet, go, rustup, gem, and ghcup are documented throughout the language reference files in the references/ directory.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted project files and configurations which are then processed by analysis scripts.
  • Ingestion points: scripts/detect-lsp.ts reads file names and content from the target project directory.
  • Boundary markers: None identified in the scanning scripts.
  • Capability inventory: The skill can execute various system commands and perform dynamic code loading.
  • Sanitization: The detect-lsp.ts script parses JSON configuration files.
Recommendations
  • HIGH: Downloads and executes remote code from: https://bun.sh/install - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
Sep 8, 2026, 05:15 AM