Claude Code Slash Commands · page 5

Write or update Playwright E2E tests following project conventions

Analyze PostHog usage data with correct filters, timelines, and caveats.

Comment out inappropriate logging before commit

Build, refine, and publish cumulative public release notes, then promote the current alpha prerelease to stable

Pull open GitHub PRs into the github-pr tracker as backlog items

Git pull, resolve conflicts, push, fix hook errors, then release

Refactor CLAUDE.md by extracting sections to dedicated docs with path-scoped rules

Prepare and execute an alpha release (patch/minor/major)

Release a new version of a marketplace extension

Restart the Nimbalyst application

Review all changes in current branch (read-only analysis)

Draft a response to a user reaching out on Discord or another social channel

Author a new contextual tip for Nimbalyst's tips system (transcript-inline tip cards with trigger/targeting rules).

Add help content tooltips to existing UI elements.

Create a tracking item in the Nimbalyst tracker system.

Update Anthropic Agent SDK, MCP library, and Codex SDK to latest versions

Create a guided walkthrough for a feature or workflow.

Generate comprehensive test coverage for existing code.

Reflect on session corrections and update CLAUDE.md (with human review)

Discover skill candidates from repeating session patterns

Discard queued learnings without processing

View the learnings queue without processing

Apply labels to GitHub issues

Integrate a spec-driven/skill framework as a bundled agtx plugin. Pass a GitHub repo URL to auto-generate the plugin.toml, or run without arguments to write one from scratch.

AI DevKit · Compare implementation with design and requirements docs to ensure alignment.

AI DevKit · Pre-push code review against design docs.

AI DevKit · Execute a feature plan task by task.

AI DevKit · Scaffold feature documentation from requirements through planning.

AI DevKit · Store reusable guidance in the knowledge memory service.

AI DevKit · Review feature design for completeness.

AI DevKit · Review feature requirements for completeness.

AI DevKit · Update planning docs to reflect implementation progress.

AI DevKit · Add tests for a new feature.

Agentlas commands — one entry point for network, build, storm, call, and the rest.

Use the Agentlas browser hardpoint for browser-required work.

Prepare explicitly named Agentlas Hub or Cloud agents.

Connect Agentlas agents or teams to Telegram.

Staff a task only from public Agentlas Hub agents.

Staff a task only from Agentlas agents registered on this machine.

Sign this machine into Agentlas (opens the browser sign-in window).

Turn the persistent Agentlas One personal agent on or off, or rename it.

Set or show which model runs the orchestrator and which runs the workers.

Search Agentlas Cloud and Hub agent candidates without invoking.

Update the installed Agentlas runtime and every host adapter on this machine.

Use the Agentlas browser hardpoint for browser-required work.

Build, repair, or package Agentlas agents and teams with Hephaestus.

Prepare explicitly named Agentlas Hub or Cloud agents.

Staff a task only from the signed-in owner's Agent Cloud agents.

Connect Agentlas agents or teams to Telegram.

Build an Agentlas automation by describing it, list saved ones, or request a run.

Staff a task only from public Agentlas Hub agents.

Staff a task only from Agentlas agents registered on this machine.

Sign this machine into Agentlas (opens the browser sign-in window).

Staff a task from registered Local, owner Cloud, and public Hub agents.

Set or show which model runs the orchestrator and which runs the workers.

Search Agentlas Cloud and Hub agent candidates without invoking.

Run a force-robust Stormbreaker loop — route to real agents, execute a verified pipeline to completion.

Update the installed Agentlas runtime and every host adapter on this machine.

Publish an Agentlas package to Agent Cloud or the public Hub.

Run a Brooks-Lint test quality review

Workflow recipe — the org-side AI adoption arc, policy to proof, by chaining 4 skills.

Set up or use your Professional Brain — init, ingest, recall, or review.

Workflow recipe — turn the quarter's raw numbers into a leadership story and board deck by chaining 4 skills.

Workflow recipe — review a design end-to-end, ending in measured numbers rather than adjectives, by chaining 4 skills.

Compress a document or update into a crisp executive summary.

Convene your standing AI staff — memos on every beat, a board session without you, minutes with dissent preserved.

Workflow recipe — turn a growth goal into a funnel diagnosis, experiment backlog, retention loop, and lifecycle journeys by chaining 4 skills.

Build a weighted customer health scorecard for an account.

Workflow recipe — go after a specific role end to end by chaining 4 skills: decode the JD, research the company, tailor your application, and prep the interview.

Workflow recipe — go from competitive landscape to a fully checklisted launch and press release by chaining 5 skills.

Workflow recipe — take an AI/LLM feature from a probabilistic-aware PRD to a launch-ready model card by chaining 5 skills.

Draft a product requirements document from a feature idea or brief.

Stress-test a plan, strategy, PRD, or launch with a room of hostile expert personas before you commit.

Workflow recipe — turn one blog post, video, or idea into a full platform-native content pack (thread, LinkedIn, newsletter, carousel, short-form script) with sharp hooks and a thumbnail concept.

Workflow recipe — diagnose an at-risk customer and build the full save play through to renewal by chaining 4 skills.

Run a structured sprint retrospective from notes.

Score and rank initiatives with the RICE framework.

Workflow recipe — go from a vague opportunity to validated insight and a prioritised next step by chaining 4 skills.

Set up your Skill Memory — a pm-context.md every skill reads so outputs come back tailored to you.

Onboard a new user — find out what they do, recommend the right bundles & top skills, and set up a project CONTEXT.md so every skill is tailored to them.

Workflow recipe — take a feature idea from fuzzy brief to launch plan by chaining 5 skills.

Workflow recipe — make your product agent-usable by chaining 4 skills, spec to pricing.

Plan a sprint with a calibrated, realistic commitment.

Add tests for recently changed files or specified code

System design and architecture planning mode. Focuses on high-level design, trade-offs, and technical decisions before implementation.

Explore the repository with 10 parallel subagents to create comprehensive documentation. Creates CODEBASE.md with full architecture analysis.

Simplifies code after implementation. Reviews recent changes and suggests/applies simplifications while preserving behavior.

Create a conventional commit with auto-generated message

Commit staged changes, push to remote, and create a pull request

Safe, one-at-a-time dependency upgrades with verification after each. Detects package manager automatically.

Runs linting and code quality checks. Catches style issues, potential bugs, and enforces project standards.

Auto-fix all linting and formatting issues

Educational mode that explains concepts, teaches patterns, and guides learning. Ideal when learning a new codebase, technology, or programming concept.

View agent performance metrics and session history. Shows usage patterns, files changed, and session summaries.

Multi-perspective analysis using parallel subagents. Analyzes code from architecture, security, performance, and testing viewpoints simultaneously.

Run parallel code review across multiple directories or files. Uses N subagents simultaneously for faster analysis.

Create and maintain a structured PLAN.md for persistent task tracking. Supports phases, dependencies, and progress tracking.

Quickly fix lint errors, type errors, or simple bugs

Fast-paced development mode for quick iterations. Minimal ceremony, maximum speed. Best for prototypes, MVPs, and time-sensitive tasks.

Systematic, safety-first refactoring with verification at each step. Never refactors and adds features simultaneously.