swift-lsp

Fail

Audited by Gen Agent Trust Hub on Apr 28, 2026

Risk Level: HIGHCOMMAND_EXECUTIONPROMPT_INJECTIONSAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions include a command requiring elevated privileges: sudo ln -s /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/sourcekit-lsp /usr/local/bin/sourcekit-lsp. Modifying system-protected paths like /usr/local/bin using sudo is a privilege escalation vector that can be abused to intercept or replace system binaries.- [PROMPT_INJECTION]: The skill analyzes Swift project files to provide diagnostics and inform the agent's editing plan. This creates an indirect prompt injection vulnerability.
  • Ingestion points: The skill reads .swift files and project metadata (Package.swift, .xcodeproj) within the workspace.
  • Boundary markers: No delimiters or instructions are present to prevent the agent from obeying instructions hidden in code comments or strings.
  • Capability inventory: The agent is authorized to edit files based on the LSP diagnostics it receives.
  • Sanitization: No validation or sanitization of the source code content is performed before it is used to influence the agent's planning phase.- [SAFE]: The skill correctly directs users to the official swift.org/install website for the Swift toolchain, which is a recognized and trusted source for these development tools.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Apr 28, 2026, 01:39 PM
Security Audit — agent-trust-hub — swift-lsp