Skill Claude CodeCodex
Compress natural language memory files (CLAUDE.md, todos, preferences) into caveman format to save input tokens. Preserves all technical substance, code, URLs, and structure. Compressed version overwrites the original file. Human-readable backup saved as FILE.original.md. Trigger: /caveman:compress or "compress memory…
Skill Claude CodeCodex
Ultra-compressed communication mode. Cuts token usage 75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or…
Skill Claude CodeCodex
Guidelines for creating formula-driven Excel workbooks with xlsxwriter. Inject live Excel formulas with cached values so workbooks display immediately and recalculate when actuaries change selections. Covers formula patterns, cross-workbook references, and shared formatting modules.
Skill Claude CodeCodex
Guide for improving GitHub Copilot Agent performance through skills, custom instructions, AGENTS.md, or other configuration. Use this when asked to improve agent behavior, create/update skills, modify custom instructions, update AGENTS.md, or customize GitHub Copilot performance in repositories.
Skill Claude CodeCodex
Guide for Python development in this project. Use any time you write Python.
Plugin Claude Code
Actuarial reserving workflow bundle for Claude Code, shipping all TeamAnalyst skills and selector agent references in one plugin.
Plugin Claude Code
TeamAnalyst actuarial reserving workflow bundle. Installs all skills and selector agent references for reserve analysis workflows.
Instructions file CodexOpenCode
Instructions for cas-team-analyst/team-analyst.
Instructions file
Instructions for cas-team-analyst/team-analyst, covering claude.md — teamanalyst, readme is a product artifact, project overview, file structure and what owns what and single source of truth files — edit these directly.
Instructions file Gemini CLI
Instructions for cas-team-analyst/team-analyst: @./skills/reserving-analysis/SKILL.md @./skills/peer-review/SKILL.md @./skills/excel-formulas/SKILL.md @./skills/python/SKILL.md @./skills/improve-agent/SKILL.md.
Skill Claude CodeCodex
Peer review of completed reserve analysis. Use after reserving-analysis workflow finishes. Do NOT use during initial analysis (separate skill exists).
Skill Claude CodeCodex
Actuarial reserving analysis workflow. Use when the user requests reserving. Do NOT use for peer review (separate skill exists).
Agent
Framework AI LDF selector for chain-ladder reserving across all measures. Applies structured decision framework with documented criteria. Invoke once to make LDF selections for all measures (Paid Loss, Incurred Loss, Reported Count, etc.) in the analysis.
Agent
Open-ended AI LDF selector for chain-ladder reserving across all measures. Makes selections using actuarial judgment and pattern recognition without a rigid rules framework. Invoke once to make LDF selections for all measures (Paid Loss, Incurred Loss, Reported Count, etc.) in the analysis.
Agent
Framework AI tail curve selector for chain-ladder reserving across all measures. Applies a phased tail curve decision framework with required documentation for ASOP 43 compliance. Invoke once to make tail curve selections for all measures in the analysis.
Agent
Open-ended AI tail curve selector using holistic actuarial judgment and pattern recognition across all measures. Makes independent tail curve selections based on curve diagnostics, triangle characteristics, and experience without rigid rule sequencing. Invoke once for all measures in the analysis.
Agent
Framework AI selector for ultimate losses and counts by accident year. Applies structured framework to weight Chain Ladder, BF, Cape Cod, Berquist-Sherman, Frequency-Severity, Benktander, and related methods based on maturity, diagnostics, and data conditions. Makes one selection for Loss (choosing between…
Agent
Open-ended AI selector for ultimate losses and counts by accident year. Makes selections using actuarial judgment and pattern recognition without a rigid rules framework. Provides creative second opinion alongside framework selector by holistically weighing method indications. Makes one selection for Loss (choosing…