Command
Translate a natural-language optimization problem into a complete, runnable discopt model — variables, objective, constraints, validation, and a solve call. Use when the user describes a problem in words and wants discopt model code.
Command
Analyze a discopt model and suggest reformulations that strengthen relaxations, tighten bounds, exploit structure, or speed up the solve — big-M, bilinear/McCormick, RLT, PSD/SOC, GDP hull, geometric programming, decomposition, FBBT. Use to make an existing model solve faster or close the gap.
Instructions file
Instructions for jkitchin/crucible, covering crucible: llm-compiled knowledge base, directory layout, conventions, org-mode articles and ingest workflow.
Settings file Claude Code
Agent settings declaring 9 allowed tools.
Skill Claude CodeCodex
LLM-compiled knowledge base management. Ingests primary sources (papers, notebooks, data), distills them into interlinked org-mode wiki articles, and provides structured search over the wiki graph database. Triggers: "ingest this", "distill", "add to crucible", "crucible ingest", "crucible search", "search the wiki"…
MCP server Claude CodeCodexCursor
MCP server "fairchem-mcp" as configured in jkitchin/fairchem-mcp. Runs locally from the fairchem-mcp Python package.