The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the mcp__lsp__execute_command tool to run code generation logic. These commands are provided directly by the underlying language server as standard code actions (e.g., 'Implement interface').
[EXTERNAL_DOWNLOADS]: The documentation references the official agent-lsp repository on GitHub (github.com/blackwell-systems/agent-lsp). This is a trusted resource from the skill's author required for the skill to function.
[PROMPT_INJECTION]: The skill processes structured code actions and diagnostics from the LSP server. While it ingests untrusted source code as part of its operation, it does not include instructions that attempt to bypass agent safety filters or override system prompts.
[SAFE]: The skill is designed for development productivity and uses vendor-provided tools for their intended purposes. It includes safety constraints such as requiring user confirmation when multiple generation options are available.

lsp-generate