Agent skill · data analytics · jeremylongshore

glean-cost-tuning

Optimize Glean costs by managing indexed content volume, datasource

Why this skill is useful

Provides specific domain knowledge and executable scripts for optimizing Glean costs that the AI wouldn't reliably generate on its own.

What it needs

About 3k tokens when loaded. Last updated 2026-08-07. 2,604 stars on the source repository.

What this skill does

Glean Cost Tuning Overview Glean pricing scales with indexed content volume and per-seat user count, making document indexing volume and search query frequency the primary cost drivers. Enterprise deployments typically connect dozens of datasources, each pushing thousands of documents into the index. Without active content governance, stale drafts, archived pages, and near-empty documents inflate the index by 30-50%, driving up costs with zero search value. Pruning irrelevant content and using incremental indexing are the highest-leverage optimizations. Cost Breakdown Component Cost Driver Optimization ----------- ------------ -------------- Document indexing Volume of indexed content across all sources Filter drafts, templates, and archived content pre-index User seats Per-seat licensing Audit active users quarterly; deprovision inactive accounts Search queries Query volume across the organization Cache frequent queries; use search analytics to identify redundant patterns Datasource connectors Number of active connectors to maintain Consolidate overlapping sources; remove unused connectors Content storage Size of indexed documents Truncate body to 50KB; skip attachments over 10MB API Call Reduction Usage Monitoring Cost Optimization Checklist [ ] Filter drafts, templates, and archived documents before indexing [ ] Prune documents not updated in 12+ months [ ] Use incremental indexing — only process changed documents [ ] Truncate document bodies to 50KB maximum [ ] Consolidate overlapping datasource connectors [ ] Audit user seats quarterly and deprovision inactive accounts [ ] Skip attachments larger than 10MB [ ] Monitor index utilization with 80% threshold alerts Error Handling Issue Cause Fix ------- ------- ----- Index bloat exceeding budget No content filtering on connectors Apply shouldIndex filter to all datasource pipelines Stale search results Deleted docs still in index Run nightly reconciliation to remove orphaned entries Connector timeouts Source system …

How to use it

Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:

@skills jeremylongshore/glean-cost-tuning--d1b7bd

View the source on GitHub

Browse the @skills marketplace