Linear

Create and manage Linear issues using the GenLayer issue template for the CI/E2E Runner project.

Overview

This skill handles Linear issue management for the CI/E2E Runner project:

• Creates issues using the standard template format
• Converts plans/reports into properly formatted Linear issues
• Supports creating issues in backlog or specific cycles

Configuration

• Team: Node (NOD) -- default; configurable per organization needs
• Default State: Backlog (unless cycle specified)

Issue Template

All issues are created with this structure:

## Problem Statement

[What problem are we solving and why?]

## Proposed Solution

[High-level approach]

## Acceptance Criteria

1. Given X, when Y, then Z (behavior specifications)
2. [Performance: <100ms response (if applicable)]
3. [Testing requirements (if applicable)]
4. [AI-generated code reviewed by human requirements (if applicable)]
5. [Edge cases considered beyond AI suggestions (if applicable)]

## Specific AI-Execution Plan

\`\`\`
[What parts will be solved by the AIs above]
\`\`\`

## Human Contribution Focus

[What parts will require human creativity/decision-making?]

## Technical Notes

[Implementation details, gotchas, AI limitations encountered]

## Lessons Learned *(optional)*

[What was learned while planning the ticket?]

Note: "Lessons Learned" is optional at creation. Include if there are insights discovered while planning. Omit section if empty.

Content Extraction from Plans

When given a plan/analysis/report, extract content using these mappings:

Priority when content fits multiple sections:

1. problem_statement = WHY (motivation, pain)
2. proposed_solution = WHAT (high-level approach)
3. ai_execution_plan = HOW AI does it (specific steps)
4. acceptance_criteria = HOW we verify (success criteria)
5. human_contribution = WHAT humans decide
6. technical_notes = WHAT could go wrong

Empty sections: Omit entirely. Never use placeholder text.

AI-Execution Plan Format

The AI-Execution Plan must be actionable steps another Claude agent can execute:

## AI Execution Steps

### Step 1: [Action] [What] in [Where]
- File: path/to/file.ts
- Pattern: [what to look for]
- Action: [Add/Remove/Update] [specific change]
- Verify: [how to confirm done]

### Step 2: ...

Each step needs:

• File/Pattern: Where to make changes
• Action: Specific change to make
• Verify: How to confirm it's done

Proposal Wizard

Before creating a ticket, show a proposal for discussion:

## Ticket Proposal

**Title:** [inferred title]

### Metadata
| Field | Value | Reasoning |
|-------|-------|-----------|
| Labels | Productivity | CI/CD tooling improvement |
| Estimate | 2pt | 3 files, moderate changes, ~1 day |
| Cycle | next | User specified |
| State | Todo | Auto (cycle specified) |
| Related | NOD-500 | Found in plan references |

### Description Preview
[First lines of each section...]

---
**Ready to create?** Or adjust any options?

Wait for user approval before creating.

CRITICAL: Re-approval Required After Changes

When user requests ANY changes (even simple ones like "assign to X"):

1. Apply the requested changes
2. Show the FULL updated proposal again
3. Wait for EXPLICIT approval ("yes", "create it", "looks good")
4. NEVER create the ticket immediately after changes are requested

Available Labels

Metadata Inference

Estimate Guide (Total Effort = AI + Human Review + Iterations)

Complexity multipliers (+1 point each):

• Touches CI pipeline / critical workflow path
• Has risks/edge cases listed
• Requires testing strategy
• Cross-module changes

Quick Reference

Creates

• Linear issues in the configured team with proper template formatting

Required Fields

• title - Issue title
• problem_statement - What problem are we solving?

Optional Template Fields

• proposed_solution, acceptance_criteria, ai_execution_plan, human_contribution, technical_notes

Optional Metadata

Optional Relationships

Usage

When invoked directly by the user:

1. Request required information if not provided
2. Format content using the template
3. Create issue in Linear with appropriate metadata

Full Documentation

See skill.yaml for complete procedure, patterns, and template details.