linkedin-skills
sergebulaev/linkedin-skills is an open-source (MIT licensed) collection of 12 modular agent skills specifically engineered for Claude Code, Codex, Claude Desktop, OpenClaw, and Hermes Agent. Designed to counter generic, mechanical LLM marketing copy, it enforces a strict human-in-the-loop workflow where drafts require explicit user approval before publishing.
The skills span the complete LinkedIn content lifecycle: viral hook reverse-engineering, profile optimization for inbound B2B lead generation, 7-day content planning, and context-aware post commenting. Crucially, the toolkit incorporates dedicated anti-AI auditing skills that strip telltale synthetic markers, repetitive cliché transitions, and robotic phrasing to ensure posts read genuinely.
Adhering to the agentskills.io SKILL.md specification, the package installs cleanly via CLI or direct git clone, with symlinks mapping seamlessly to Claude's runtime. However, live URL inspection and automated post scheduling require pairing with third-party scrapers (such as Apify) and webhook dispatchers, and users must exercise caution to avoid triggering LinkedIn anti-automation detection.