Agentica server + Claude proxy setup - architecture, startup sequence, debugging

Meta-skill for creating and managing Claude Code skills

Environment Triage

Create and use Claude Code slash commands - quick prompts, bash execution, file references

Skill Development Rules

TDD workflow for migrations - orchestrate agents, zero main context growth

Parallel Agent Type Contracts

Braintrust tracing for Claude Code - hook architecture, sub-agent correlation, debugging

Create handoff document for transferring work to another session

AI-powered web search, research, and reasoning via Perplexity

Show users how Continuous Claude works - the opinionated setup with hooks, memory, and coordination

Maps questions to the optimal tldr command. Use this to pick the right layer

Code quality checks, formatting, and metrics via qlty CLI

Completion Check: Verify Infrastructure Is Wired

Meta-skill for internal codebase exploration at varying depths (quick/deep/architecture)

System health check (MOT) for skills, agents, hooks, and memory

Idempotent Redundancy

Document codebase as-is with thoughts directory for historical context

Graceful Degradation with Helpful Messages

Git Commit Rules