linkedin-browser-research - SKILL.md Agent Skill

name: linkedin-browser-research description: Use browser-backed LinkedIn adapters to inspect company pages, people profiles, hiring signals, and professional discussion context around industries, products, and firms. metadata: {"marketbot":{"emoji":"💼","triggers":["linkedin","linkedin profile","hiring signal","company page","professional discussion"],"output":"linkedin-browser-research-report","risk":"low","freshness":"live","tools":["browser_site"],"required_tools":["browser_site"],"markets":["global","mixed"],"asset_classes":["equity","macro","etf"],"task_type":"browser-research","determinism":"tool-backed","priority":78}}

Use this skill when the user wants professional-network context such as hiring signals, company activity, or profile-based ecosystem clues.

Use browser_site with LinkedIn adapters that exist in the runtime catalog. Prefer exact adapters such as:
- linkedin/search
- linkedin/profile
Read references/adapter-examples.md when you need concrete adapter call patterns or fallback behavior.
Focus on:
- hiring intensity
- company and team activity
- industry positioning and ecosystem signals
Pair with github-browser-research when both professional and open-source traction matter.

Do not invent undocumented linkedin/* adapters. If the runtime catalog does not expose the one you need, say so and continue with the closest listed adapter.
Treat LinkedIn as professional-signal context, not as audited business data.
Separate profile claims from independently verified facts.