glean-core

Configure a Glean MCP server connection

Check Glean MCP server connection status

Use when you need to communicate the reliability, freshness, or authority of information from Glean. Triggers when presenting search results, when data might be stale, when sources have different authority levels, when the user should verify information, or when you're uncertain about completeness of results.

Use when the user asks about company documents, internal wikis, policies, specifications, design docs, RFCs, or enterprise knowledge. Triggers on phrases like "find the doc about", "what's our policy on", "where is the spec for", "company guidelines", "internal documentation", or when searching for information that would be in enterprise systems rather than the local codebase.

Use when Glean MCP tools are available and you need guidance on which tool to use, how to format queries, or best practices for enterprise search. This skill provides tool selection logic and query optimization for Glean integrations. Auto-triggers when mcp__glean tools are being considered.

Use when the user asks about meetings, meeting history, decisions made in meetings, action items from discussions, or what was discussed. Triggers on phrases like "what was decided", "meeting about", "action items from", "what happened in the meeting", "meeting notes", "transcript", "who attended", or when needing context from past conversations and discussions.

Use when the user asks about people, employees, team members, org structure, or expertise. Triggers on phrases like "who works on", "who is responsible for", "who owns", "find someone who knows", "who should I talk to", "who reports to", "team members", "org chart", or any question starting with "who" about the company.

Use when combining information from multiple Glean sources or when needing to synthesize results across documents, meetings, code, and people searches. Triggers on complex queries that span multiple data types, when results seem contradictory, when building comprehensive answers from partial information, or when the user asks for a complete picture of something that requires multiple queries.

1 hook across 1 event