Discover and install specialized agent skills from the open ecosystem when users need extended capabilities. Helps identify relevant skills by domain and task when users ask "how do I do X" or "find a skill for X" Provides npx skills find [query] command to search interactively and npx skills add <package> to install from GitHub or other sources Recommends skills based on install count (prefer 1K+), source reputation (official sources like Vercel, Anthropic), and GitHub stars to ensure quality Checks the skills.sh leaderboard first for battle-tested, popular options before running CLI searches
Relentless interviewing skill that stress-tests plans and designs through systematic questioning. Conducts deep-dive questioning across all aspects of a plan, walking through decision trees branch-by-branch until shared understanding is reached Automatically explores the codebase to answer questions where code context is available, reducing redundant back-and-forth Designed for design reviews, architecture validation, and pre-implementation planning where thorough vetting prevents downstream issues
Relentless design interrogation that stress-tests plans against your domain model and sharpens terminology inline. Explores your codebase to ground discussions in existing code, glossaries (CONTEXT.md), and architectural decisions (ADRs) Challenges fuzzy language, resolves terminology conflicts, and updates your domain glossary as decisions crystallize Tests design decisions against concrete scenarios to expose edge cases and boundary violations Creates or updates CONTEXT.md and ADRs only when warranted, keeping documentation lean and decision-focused
Analyze codebases for architectural friction and propose module-deepening refactors as testability improvements. Explores codebases organically to surface shallow modules, tightly-coupled components, and untested seams rather than following rigid heuristics Applies John Ousterhout's "deep module" principle: small interfaces hiding large implementations for better testability and AI navigability Generates multiple radically different interface designs (minimalist, flexible, caller-optimized, ports & adapters) via parallel sub-agents, then recommends the strongest approach Creates GitHub issue RFCs documenting the problem space, design trade-offs, and refactoring rationale
Fast, persistent browser automation with session continuity across sequential agent commands. Supports three browser modes: headless Chromium, real Chrome with profile support, and cloud-hosted remote browsers with proxy configuration Includes 15+ command categories covering navigation, page inspection, interactions, data extraction, cookie management, and JavaScript execution Offers cloud session management, local server tunneling via Cloudflare, and parallel subagent execution through remote sessions Built-in Python integration for setting variables, accessing the browser object, and running scripts within the automation context
Test-driven development with vertical slices, behavior-focused tests, and incremental red-green-refactor cycles. Emphasizes integration-style tests that verify behavior through public APIs, not implementation details; tests should survive refactors unchanged Requires vertical slicing (one test → one implementation → repeat) instead of horizontal slicing (all tests first, then all code), preventing brittle, behavior-insensitive test suites Includes planning phase to confirm interface changes, prioritize behaviors to test, and design for testability before writing code Provides refactoring guidelines covering duplication extraction, module deepening, and SOLID principles, applied only after all tests pass
Configuration scaffold for engineering skills to locate issue tracker, triage labels, and domain documentation. Explores the current repo to detect GitHub/GitLab remotes, existing AGENTS.md or CLAUDE.md , and domain doc layout before prompting Walks through three guided decisions one at a time: issue tracker location (GitHub, GitLab, local markdown, or custom), triage label vocabulary mapping, and domain doc structure (single or multi-context) Writes an ## Agent skills block to the appropriate root file and creates three reference docs under docs/agents/ that downstream skills ( to-issues , triage , diagnose , tdd , improve-codebase-architecture ) will read Allows editing of the draft configuration before committing changes; re-running is only needed if switching trackers or restarting from scratch
State machine-driven issue triage with role-based workflow management and agent brief generation. Two category roles ( bug , enhancement ) and five state roles ( needs-triage , needs-info , ready-for-agent , ready-for-human , wontfix ) guide issues through a standardized workflow Supports querying unlabeled issues, in-progress triage, and issues awaiting reporter feedback; presents results in priority buckets with one-line summaries Includes reproduction testing for bugs, grilling sessions for incomplete issues, and templates for triage notes and agent briefs Respects prior triage context, codebase ADRs, and out-of-scope knowledge bases to avoid re-asking resolved questions and surfacing prior rejections
Throwaway code that answers a single design or logic question through an interactive prototype. Choose between two branches: a terminal app for testing state machines and business logic, or multiple toggleable UI variations on a single route for design exploration Minimal setup with one command to run; state lives in memory with no persistence unless the question explicitly requires it Surface full state after every action so changes are immediately visible and the prototype's answer is clear Delete or fold validated decisions back into production code once the question is answered; capture the verdict before cleanup
CLI toolkit for scaffolding, validating, previewing, and rendering HyperFrames video compositions. Scaffolds projects with init (9 templates available), lints compositions for structural errors, and visually inspects layouts for text overflow and clipping issues across the timeline Preview mode hot-reloads on file changes; render supports quality tiers (draft/standard/high), multiple formats (MP4/WebM), frame rates, and parametrized variable overrides Includes environment troubleshooting commands ( doctor , browser , info ) and requires Node.js >= 22 and FFmpeg For audio preprocessing (TTS, transcription, background removal), use the hyperframes-media skill instead
Build and sharpen a project's domain model. Use when discussing codebase terminology, writing or editing a GLOSSARY.md, or recording or editing an ADR.
Shared vocabulary for designing deep modules. Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a…
Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.
Review the changes since a fixed point (commit, branch, tag, or merge-base) along two axes: Standards (does the code follow this repo's documented coding…
Implement a piece of work based on a spec or set of tickets.
Ask which skill or flow fits your situation. A router over the skills in this repo.
Systematic diagnosis and remediation for Azure production issues using AppLens, Monitor, and resource health. Covers Container Apps, Function Apps, and AKS troubleshooting with service-specific guides for image pulls, cold starts, health probes, invocation failures, and node/pod issues Integrates AppLens (MCP) for AI-powered root cause analysis and Azure Monitor (MCP) for KQL-based log and metric queries Provides a five-step diagnostic flow: identify symptoms, check resource health, review logs, analyze metrics, and investigate recent changes Includes quick reference commands for activity logs, container logs, and App Insights queries, plus routing guidance for AKS-specific incidents
Prepare Azure applications for deployment with infrastructure-as-code, configuration, and containerization. Generates Bicep or Terraform infrastructure code, azure.yaml configuration, and Dockerfiles based on application type and deployment target Requires mandatory plan-first workflow: analyze workspace, gather requirements, select recipe (AZD, AZCLI, Bicep, or Terraform), design architecture, then execute only after user approval Covers new app creation, modernization, and updates; routes specialized requests (copilot SDK, cloud migration, AI gateway) to dedicated skills before resuming preparation Hands off to azure-validate for infrastructure validation and azure-deploy for deployment execution; plan status must be updated to "Ready for Validation" before validation begins
Install the CLI. Run your first Skill in 30 seconds. Scale when you're ready.