The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references the vendor's official MCP server endpoint at https://mcp.liftli.ai/mcp and provides documentation links to liftli.ai for extended functionality.
[COMMAND_EXECUTION]: The instructions include CLI commands for the user to manually install related skills (npx skills add liftli-ai/linkedin-agent-skills) and configure the Model Context Protocol (claude mcp add). These are provided as setup instructions for the user rather than automated executions by the agent.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process potentially untrusted external data to perform its primary function.
Ingestion points: Reads user-provided LinkedIn post drafts and local voice configuration files located at .liftli/voice.md and ~/.liftli/voice.md (SKILL.md).
Boundary markers: The instructions do not specify explicit delimiters or "ignore" instructions to separate processed text from agent instructions.
Capability inventory: The skill notes potential interaction with the strategist_briefing and linkedin-publish tools if they are available in the environment.
Sanitization: No specific sanitization or validation logic is described for the processed text.

linkedin-post-rewriter