The Agent Skills Directory

[SAFE]: No security issues were detected. The skill is designed for read-only code exploration and does not perform any sensitive file modifications, network exfiltration, or privileged command execution.
[INDIRECT_PROMPT_INJECTION]: The skill has a potential attack surface by ingesting external data (source code documentation and hover text via LSP tools). However, as the skill is strictly read-only and lacks capabilities for command execution, network operations, or file writing, the risk of an indirect injection impacting the system is negligible.
Ingestion points: The hover results and documentation comments returned by mcp__lsp__inspect_symbol in SKILL.md.
Boundary markers: The results are placed within a structured "Explore Report" format.
Capability inventory: The skill is restricted to read-only LSP queries; it contains no subprocess calls, network exfiltration tools, or file-writing logic.
Sanitization: None documented for the LSP response text.

lsp-explore