Operate a locally deployed Omnigraph graph database. Use this skill whenever you see Omnigraph CLI commands (omnigraph init/read/change/load/schema/embed/branch/commit/run), .pg schema files, .gq query files, RustFS S3 URIs (s3://omnigraph-local/...), or work inside a folder containing omnigraph.yaml. Covers local RustFS setup, project layout, schema authoring and evolution (plan before apply), query linting, data changes (change vs load --mode merge vs overwrite), branches for data review, embeddings, aliases for automation, HTTP server operation, Cedar policy, and common gotchas. Especially important BEFORE running schema apply (plan first), any load (pick --mode carefully), or any .gq/.pg edit (lint afterward). Apply this skill aggressively when the user mentions Omnigraph, graph migrations, or graph database local development.
Bootstrap a new Omnigraph-based SPIKE industry intelligence graph from scratch. Use this skill whenever a user wants to set up a new SPIKE graph — either with the existing AI industry demo data or for a new domain (biotech, fintech, crypto, geopolitics, macroeconomics, SaaS, climate tech, etc.). The flow presents a demo-vs-custom decision, then for custom setups asks about domain scope, actors, cadence, and sources, adapts schema and enums for the target domain, runs initial web research to generate real seed content, and executes init + load. Apply aggressively when the user says any of: set up Omnigraph, bootstrap a new graph, create a new SPIKE starter, I want to track X industry, initialize intel for Y, new graph for Z domain, start a new context graph, or similar phrasing. This skill takes a user from zero to a populated, queryable graph.