resolve-slate-issue

Resolve Slate GitHub issues end to end with verified fixes and issue comments.

16.5k|997|Updated Dec 2, 2019
One-click install
npx skills add https://github.com/udecode/plate --skill resolve-slate-issue
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: resolve-slate-issue
Source: https://github.com/udecode/plate/tree/main/.agents/skills/resolve-slate-issue
Command: npx skills add https://github.com/udecode/plate --skill resolve-slate-issue

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Triaging and fixing Slate or Slate v2 GitHub issues manually requires reproducing bugs, verifying fixes against current source, and writing accurate issue comments, which is slow and error-prone.

Core Features & Use Cases

  • Issue Intake and Classification: Fetches the issue via the GitHub CLI, extracts the user flow, expected and actual behavior, and classifies the state as red-current, already-accounted, needs-manual-proof, or blocked.
  • Proof-Driven Fixing: Adds the smallest failing test or Playwright row, fixes the durable owner in the Slate v2 source, and runs focused verification commands.
  • Verified Issue Comments: Posts a structured confidence-rated comment with gh and closes the issue only when verification reaches 95% or higher confidence.
  • Use Case: A maintainer receives a Slate editor bug report with a screen recording; the Skill transcribes the video, reproduces the flow against current source, fixes the owning package, and posts a verified comment.

Quick Start

Resolve Slate issue number 123 from the udecode/slate repository and post the verified result as an issue comment.

Frequently Asked Questions about resolve-slate-issue

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

FAQPage Schema
How do I resolve a GitHub issue end to end with an AI agent?

Provide the issue number or URL, and the Skill fetches it with gh, reproduces the reported flow against current source, fixes the owning code, verifies with tests, and posts a structured comment. It closes the issue only at 95% or higher confidence.

How does the Skill verify a Slate bug fix before commenting?

It adds the smallest failing proof such as a package test or Playwright row, fixes the durable owner in .tmp/slate-v2, then runs focused commands like bun typecheck, bun check, or Playwright against the issue flow.

Does the Skill create pull requests for issue fixes?

No. The Skill explicitly forbids opening, updating, or mentioning pull requests and forbids PR metadata lines in comments. It only posts a verified issue comment and optionally closes the issue.

What happens if the issue is already fixed in current Slate v2 source?

The Skill classifies it as already-accounted, makes no code change, verifies current behavior with the smallest honest command, and comments that the flow already passes. It closes the issue only at 95% or higher confidence.

Can the Skill handle issues requiring manual device or IME testing?

It classifies these as needs-manual-proof, runs any honest supporting automation, and comments with an exact human reproduction request. The issue stays open and the confidence line is not used.