Commands Overview
Hyprlayer installs slash commands that you invoke from your AI tool (Claude Code, GitHub Copilot, or OpenCode). These commands orchestrate a spec-driven development workflow.
Why Commands?
Section titled “Why Commands?”Claude Code, GitHub Copilot, and OpenCode are powerful on their own. They can read your codebase, edit files, run commands, and spawn sub-agents. If your task is simple, you don’t need commands — just ask.
The problems show up when the work is non-trivial. Ask any of these tools to “build feature X” and they start writing code immediately. Research, planning, and implementation all happen in one unstructured pass. There’s no checkpoint where you review a plan before code gets written. Context evaporates between sessions. Commit messages are generic. The model agrees with everything you say.
Commands fix this by encoding enforced phase separation (research, plan, implement, and validate are distinct steps with artifacts between them), a persistent shared knowledge base (the thoughts directory gives AI agents team-wide memory that survives across sessions and tools), specialized sub-agents with focused roles (not generic helpers, but domain-specific agents that understand your codebase patterns, thoughts directory, JIRA instance, and the web), automatic model selection (Opus for research and planning, Sonnet for execution), and behavioral constraints you’d forget to specify every time (skepticism during planning, no AI attribution in commits, template adherence for PRs).
Each command page below explains what that specific command adds beyond what the base tools do.
When to Skip Commands
Section titled “When to Skip Commands”Commands aren’t always the right choice:
- Quick one-off questions — “What does this function do?” doesn’t need
/research_codebase - Simple changes — Renaming a variable or fixing a typo doesn’t need a plan
- Exploration — Sometimes you want to think out loud with the model before committing to a structured workflow
The commands are designed for non-trivial work where the structure pays for itself. Use your judgment.
Core Workflow Commands
Section titled “Core Workflow Commands”| Command | Description | Model |
|---|---|---|
/research_codebase | Document how existing code works | Opus |
/create_plan | Create a phased implementation plan | Opus |
/iterate_plan | Refine an existing plan | Opus |
/implement_plan | Execute a plan phase-by-phase | Sonnet |
/validate_plan | Verify implementation against plan | Sonnet |
Git & Shipping Commands
Section titled “Git & Shipping Commands”| Command | Description | Model |
|---|---|---|
/commit | Create git commits with user approval | Sonnet |
/describe_pr | Generate PR description from template | Sonnet |
Collaboration Commands
Section titled “Collaboration Commands”| Command | Description | Model |
|---|---|---|
/create_handoff | Write a handoff document for another session | Sonnet |
/resume_handoff | Pick up work from a handoff document | Sonnet |
/local_review | Set up a worktree to review a branch | Sonnet |
Utility Commands
Section titled “Utility Commands”| Command | Description | Model |
|---|---|---|
/founder_mode | Retroactively create JIRA ticket and PR | Sonnet |
/cost_estimate | Estimate development costs | Sonnet |
Model Selection
Section titled “Model Selection”Commands that require deep reasoning and research use Opus (Claude’s most capable model). All other commands use Sonnet for faster execution. When using OpenCode, the specific model IDs depend on your configured provider.
This isn’t just about cost optimization. Opus and Sonnet have genuinely different strengths. Using Opus for research and planning produces more thorough analysis. Using Sonnet for implementation produces faster iteration cycles. Commands make this selection automatically — you don’t have to think about which model fits the task.
Cross-Tool Portability
Section titled “Cross-Tool Portability”The same commands and agents are installed for whichever tool you’re using — Claude Code, GitHub Copilot, or OpenCode. The thoughts directory is tool-agnostic. A plan created with Claude Code can be implemented with Copilot. A handoff written in one tool can be resumed in another. The workflow doesn’t lock you into a specific AI tool.