garura:capture

Convert structured issue inputs into type-driven GitHub issues.

3|Updated Jan 25, 2026
One-click install
npx skills add https://github.com/kapilvirenahuja/garura --skill garura-capture
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: garura:capture
Source: https://github.com/kapilvirenahuja/garura/tree/main/core/components/plays/capture
Command: npx skills add https://github.com/kapilvirenahuja/garura --skill garura-capture

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill turns messy issue intake into consistent, type-driven GitHub issues that can be filed asynchronously without blocking the caller.

Core Features & Use Cases

  • Structured capture intake: Accepts a title and problem statement, optionally including evidence, root cause, specific issues, expected behavior, and impact.
  • Type inference and validation: Uses an explicit type when provided, otherwise infers the issue type from problem text, with special handling for defect severity.
  • Deterministic issue filing: Formats GitHub issue titles and bodies with consistent prefixes and a metadata table, applies/creates the required label, and dispatches filing as a background task with a local fallback when GitHub is unreachable.
  • Use case: Capture a production incident report from an agent and file a labeled “defect” (with required severity) so engineering can triage quickly without waiting on GitHub network calls.

Quick Start

Ask the AI to capture your issue with a title and problem, optionally providing type and severity, and it will dispatch a background GitHub issue filing confirmation.

Frequently Asked Questions about garura:capture

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

FAQPage Schema
How do I automatically file structured GitHub issues from async agent inputs?

To automatically file structured GitHub issues, this Skill converts title and problem inputs into type-driven markdown issue bodies and dispatches them asynchronously. It formats metadata tables, applies required labels, and handles network failures with a local fallback.

Can I infer GitHub issue types from problem text instead of providing explicit labels?

Yes, you can infer GitHub issue types from problem text. When an explicit type is not provided, the Skill infers the issue type from the problem statement, applying special handling rules for defect severity and resolving ambiguous inferences.

What is the best way to capture production incident reports and file them as labeled defects?

The best way to capture production incident reports is providing a title and problem statement to this Skill. It infers the defect type, enforces required severity fields, formats a structured markdown body, and dispatches a labeled GitHub issue asynchronously.

How does fallback handling work when GitHub is unreachable during issue dispatch?

Fallback handling during issue dispatch activates when GitHub is unreachable. The Skill creates a portable local fallback file containing the formatted issue data, ensuring no structured capture inputs are lost during asynchronous network failures.

Do I need to provide evidence and root cause sections to file a GitHub issue?

You do not need to provide evidence and root cause sections to file a GitHub issue. The Skill requires only a title and problem statement, while evidence, root cause, specific issues, expected behavior, and impact are optional inputs.

Why does async dispatch matter for GitHub issue filing?

Async dispatch matters for GitHub issue filing because it prevents blocking the caller during network requests. This allows agents or human callers to capture and submit structured issues immediately without waiting for GitHub API responses.