Agent skill · data analytics · jeremylongshore
juicebox-core-workflow-b
Execute Juicebox enrichment and outreach workflow.
Why this skill is useful
Adds specific commands and workflows for analyzing candidate data that the AI wouldn't reliably generate on its own.
What it needs
Requires npm installed locally. Requires juicebox account access. About 2k tokens when loaded. Last updated 2026-08-07. 2,604 stars on the source repository.
What this skill does
Juicebox — Advanced Analysis Overview Build custom queries, apply multi-dimensional filters, and run cross-dataset analysis on your Juicebox people-intelligence data. Use this workflow when you need to go beyond standard search — comparing candidate pools across roles, analyzing skill density by geography, or identifying talent trends over time. This is the secondary workflow; for basic search and enrichment, see juicebox-core-workflow-a. Instructions Step 1: Build a Custom Query with Filters Step 2: Run Cross-Dataset Comparison Step 3: Aggregate Skill Density by Region Step 4: Export Analysis Results Error Handling Issue Cause Fix ------- ------- ----- 400 Invalid filter Unsupported operator for field type Check field schema with client.schema.fields() 404 Dataset not found Stale dataset ID or typo List datasets with client.datasets.list() 408 Query timeout Too many filters on large dataset Add limit or narrow date range 429 Rate limited Exceeded analysis quota Implement backoff; check plan limits Partial comparison data One dataset has sparse coverage Expected — use includenulls: true for completeness Output A successful workflow produces filtered candidate lists with relevance scores, cross-dataset comparison tables showing talent market shifts, and regional skill-density rankings. Results can be exported as CSV for downstream reporting. Prerequisites An approved analysis purpose, sandbox datasets containing only synthetic records, source/destination allowlists, a suppression check, and a named owner for review and rollback. Examples Run the comparison in workspace=ci-synthetic, restrict output to aggregate metrics, verify suppression=pass; contactsexported=0, then delete the staged dataset after the redacted receipt is approved. Resources Juicebox API Docs Next Steps See juicebox-sdk-patterns for authentication and query builder helpers.
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills jeremylongshore/juicebox-core-workflow-b