What problem does it solve? Setting up a searchable local index of your LinkedIn network requires many coordinated steps — installing updates, authenticating with Powerset or a custom Modal workspace, importing and enriching a Connections.csv export, merging sources, indexing, and validating the result. This Skill turns that into a single deterministic, rerunnable nine-step checklist. ## Core Features & Use Cases - Deterministic nine-step checklist: Runs update, credential selection, Powerset login, Modal key provisioning, LinkedIn import, source merge, Modal indexing, validation, and follow-up suggestions in a fixed order with idempotent reruns. - Modal-based enrichment and indexing: Enriches LinkedIn connections on Modal sandboxes using shared caches, then builds and downloads a local DuckDB search index with runtime estimates based on connection count. - Flexible credential routes: Supports a provisioned Powerset account or a custom Modal workspace that already contains the required powerset-openai and powerset-api secrets. - Use Case: A recruiter exports their LinkedIn Connections.csv and asks their coding agent to run $setup; the agent logs into Powerset, imports and enriches 5,000 connections on Modal, merges them, builds the search index, and validates it — all in one guided session. ## Quick Start Ask your agent to run $setup and provide the path to your exported LinkedIn Connections.csv file when prompted.
