verify

Run test suites, compile checks, and behavior-anchor validation with timeouts.

8|5|Updated Feb 18, 2026
One-click install
npx skills add https://github.com/shenxingy/Clade --skill verify-shenxingy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: verify
Source: https://github.com/shenxingy/Clade/tree/main/configs/skills/verify
Command: npx skills add https://github.com/shenxingy/Clade --skill verify-shenxingy

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It ensures a codebase still behaves correctly after autonomous changes by running verification checks and re-validating documented behavior anchors.

Core Features & Use Cases

  • Project-type detection: Reads CLAUDE.md for test/verify commands and falls back to scanning common build/test indicators.
  • Test, compile, and smoke verification: Runs an appropriate test suite (with timeouts), compile/type checks, and optional verify commands.
  • Behavior anchor validation: Re-checks feature anchors defined in CLAUDE.md, marking items unverifiable when they require unavailable external dependencies.
  • Optional frontend UI interaction: For web-fullstack projects with Playwright MCP tools, it performs limited page navigation and interaction checks.

Quick Start

Ask your autonomous loop to run the verify skill after it finishes coding iterations to confirm tests, builds, and behavior anchors still pass.

Frequently Asked Questions about verify

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I verify code changes and run CI validation tests automatically?

Behavior anchor validation re-checks documented feature behaviors in CLAUDE.md after code modifications. It marks items unverifiable when they require unavailable external dependencies, ensuring documented integration points remain intact throughout autonomous runs.

Does frontend smoke testing work with web-fullstack projects?

Frontend smoke testing works with web-fullstack projects when Playwright MCP tools are available. It performs limited page navigation and interaction checks to validate core UI behaviors alongside standard test suites and compile type checks.

How do I run bounded timeout test suites after autonomous code changes?

Running bounded timeout test suites after autonomous code changes requires reading CLAUDE.md for test or verify commands, falling back to project metadata scanning, and executing targeted commands safely to produce a structured pass, partial, or fail outcome report.

What are the limitations of behavior anchor validation when external dependencies are missing?

Behavior anchor validation cannot fully verify features when external dependencies are missing, marking those items as unverifiable. It relies on CLAUDE.md anchors and project metadata, bounding its scope to available test suites, compile checks, and optional smoke commands.