GUIDE · DEVELOPMENT
A development stack should make review, testing, browser evidence, and delivery clearer—not hide more work behind a larger toolchain. Start with the weakest part of your current loop.
This guide brings together source-linked Skills for coding, testing, browser workflows, and delivery. It is intentionally smaller than the full development directory.
SELECTION ORDER
Treat each Skill as one part of an engineering loop. A stronger loop has a visible scope, an inspectable output, and a clear stop condition.
Fix one bottleneck first: plan, implementation, test, browser evidence, or release work
Prefer outputs you can inspect, such as tests, diffs, screenshots, or a written plan
Keep write and command permissions proportionate to the task
Check the source revision and first risk before combining Skills
REVIEWABLE CANDIDATES
Sorted from the same published SkillSignal metadata score used throughout the directory. Scores are signals, not safety approval.
Browser automation built for repeatable agent workflows.
Design constraints that push generated interfaces beyond generic AI UI.
Plan, implement, and evaluate high-quality MCP servers in Python or TypeScript.
Create, improve, and benchmark Agent Skills with trigger tests and comparative evaluation.
Review interfaces against practical web design and accessibility rules.
Product-specific UI design contracts grounded in real interface evidence and a hard finish gate.
DECISION BOUNDARY
A Skill that creates tests, screenshots, or plans does not prove production behavior, accessibility, security, or release readiness. Human review remains part of the loop.
KEEP EXPLORING
Open the directory to filter by Agent, operating system, price, and permission preference. The ranking is deterministic and does not call an LLM.
Open the Skill matcher ↗