SKILLSIGNAL EDITORIAL

Insights

Original, source-aware analysis for choosing agent Skills beyond raw popularity.

FEATURED REPORT

Top 100 Agent Skills in 2026: What the leaderboard actually shows

A source-aware analysis of 100 ranked entries, 92 unique Skill IDs, concentrated sources, and the limits of install popularity.

Read report ↗
PRACTICAL GUIDE

Handoff Skill: Resume Agent Work Without Repeating the Setup

When to use Matt Pocock's handoff Skill, what to save, how to find the file, and a worked example for resuming unfinished agent work.

Read report ↗
PRACTICAL GUIDE

5 Essential Agent Skills for UI/UX Designers

A source-reviewed five-role UI/UX workflow for sharper briefs, visual direction, product-specific interfaces, responsive review, and handoff evidence.

Read report ↗
PRACTICAL GUIDE

5 Agent Skills for Solo Founders Shipping an MVP

Five distinct Agent Skill roles for solo founders: challenge the idea, plan the build, create a credible interface, verify one flow, and prepare search-ready content.

Read report ↗
PRACTICAL GUIDE

Stop Installing Random Agent Skills: A 5-Step Decision Workflow

Use a five-step workflow to choose Agent Skills by task, role, source, permissions, and controlled evidence instead of popularity or collection size.

Read report ↗
PRACTICAL GUIDE

Claude Code Frontend Design Skill: One Core + Four Optional Roles

Start with Anthropic's frontend-design for visual direction. Compare optional UI and browser checks; two listed roles target Copilot, not Claude Code by default.

Read report ↗
PRACTICAL GUIDE

5 Best Agent Skills for Browser Automation and Web Testing

Compare five source-aware Agent Skills for browser control, navigation, Playwright testing, acceptance evidence, and responsive review.

Read report ↗
PRACTICAL GUIDE

5 Azure Agent Skills for Least-Privilege Cloud Operations

Compare five source-reviewed Azure Agent Skills for RBAC, deployment readiness, incident diagnosis, compute planning, and least-privilege cloud operations.

Read report ↗
PRACTICAL GUIDE

Lark and Feishu Agent Skills: Choose the Right Skill by Task

Compare Lark CLI Agent Skills for approvals, cloud files, Docs, Base, Sheets, and Wiki. Find the right task fit, dependencies, and write-risk boundaries.

Read report ↗
PRACTICAL GUIDE

Mobile Layout Skill: Fix Clipping, Spacing, and Breakpoint Failures

A task-specific workflow for using a frontend Skill to find mobile text clipping, uneven whitespace, overflow, and breakpoint failures.

Read report ↗
PRACTICAL GUIDE

Agent Browser Form Testing: A Safe Smoke-Check Workflow

A task-specific browser Skill workflow for checking form fields, validation, submission state, and evidence without touching production data.

Read report ↗
PRACTICAL GUIDE

npm Supply Chain Audit Skill: A Focused Review Workflow

A bounded workflow for reviewing npm dependencies, lifecycle scripts, lockfiles, and source changes without pretending a Skill is a security certificate.

Read report ↗
PRACTICAL GUIDE

PRD Planning Skill: Turn Requirements into Acceptance Checks

A scenario workflow for turning rough product notes into a bounded PRD with decisions, dependencies, acceptance checks, and stop conditions.

Read report ↗
PRACTICAL GUIDE

How to Install Codex Skills with npx: Project vs Personal Scope

Copy the Codex command: npx skills add <repo> --skill <name> --agent codex. Choose project or personal (-g) scope, inspect permissions, then test safely.

Read report ↗
PRACTICAL GUIDE

Code Review Skills for Claude Code: A Practical Shortlist

Compare Claude Code's official /code-review plugin with focused Agent Skills for diff analysis, static checks, browser tests, and human-reviewed evidence.

Read report ↗
PRACTICAL GUIDE

How to Check Agent Skill Permissions Before Installation

A practical permission-review method for Agent Skills covering files, shell commands, network access, accounts, destructive actions, evidence, and stop conditions.

Read report ↗
PRACTICAL GUIDE

Codex vs Claude Code Skills: Native and Portable Compatibility

Understand native versus portable Agent Skill support across Codex and Claude Code, and use a reproducible checklist before moving a workflow between Agents.

Read report ↗
PRACTICAL GUIDE

If You Could Install Only 5 Agent Skills: A Minimal Developer Stack

An opinionated five-Skill developer stack for discovery, planning, implementation support, review, and behavioral evidence—with install order and replacement rules.

Read report ↗
PRACTICAL GUIDE

The 1-3-5 Rule for Building an Agent Skill Stack

A practical rule for deciding when one, three, or five Agent Skills are enough—without creating overlapping permissions, outputs, and ownership.

Read report ↗
PRACTICAL GUIDE

5 Signs You Should Remove an Agent Skill

A practical cleanup checklist for removing Agent Skills that overlap, drift from their source, request unused access, produce ignored output, or fail controlled tests.

Read report ↗

Original, evidence-led articles now include focused scenario guides for mobile layout, browser forms, dependency risk, and PRD planning.