backlog-scan

Scan the Finans codebase against CLAUDE.md to generate a prioritized task backlog.

5|1|Updated Nov 20, 2025
One-click install
npx skills add https://github.com/andersnygaard/ettsted2 --skill backlog-scan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: backlog-scan
Source: https://github.com/andersnygaard/ettsted2/tree/main/.claude/skills/backlog-scan
Command: npx skills add https://github.com/andersnygaard/ettsted2 --skill backlog-scan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill autonomously scans the entire codebase against the CLAUDE.md specification to identify all gaps and generate a prioritized, numbered backlog of tasks. It enables planning at scale without writing code, guiding teams from broad exploration to concrete tasks.

Core Features & Use Cases

  • Comprehensive Gap Discovery: Scans the whole repo (frontend/, backend/, components/, e2e/) against CLAUDE.md to surface all gaps.
  • Prioritized Task Backlog: Produces a numbered backlog with clear task titles and priorities.
  • Planning Only: Creates tasks but does not implement or start coding.
  • Context Loading: Leverages existing rules, recent done tasks, and available skills to inform planning.
  • Use Case: When starting a new development phase or backlog is stale, run this to generate backlog and run /discover-tasks to drill into specifics.
  • Note: This skill is for planning; do not use it to implement tasks or address quick bugs.

Quick Start

Use the backlog-scan slash command to generate a comprehensive backlog. Review the created tasks and decide which to pursue. For a single specific task, use the feature-planning skill or /new-task. The following steps are provided to guide you through the backlog creation process.

Frequently Asked Questions about backlog-scan

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

FAQPage Schema
How do I generate a prioritized backlog from my codebase specifications?

Backlog-scan identifies gaps between your CLAUDE.md specification and codebase across frontend, backend, components, and e2e layers, then produces a numbered, prioritized backlog of missing features and refactors without implementing code.

Can I use backlog-scan to plan tasks without writing code?

Yes. Backlog-scan performs planning-only analysis, creating task files for your task-board via /discover-tasks while respecting the constraint that no implementation occurs.

When should I run backlog-scan in my development workflow?

Run backlog-scan when starting a new development phase or when your backlog is stale. It discovers comprehensive gaps at scale; for individual tasks, use feature-planning or /new-task instead.

What context does backlog-scan use to prioritize tasks?

Backlog-scan leverages existing rules, recently completed tasks, and available skills to inform gap analysis and task prioritization across your entire codebase.

How does backlog-scan differ from manual task creation?

Backlog-scan performs broad-to-narrow automated analysis across the full codebase to surface all specification gaps at once, whereas manual task creation addresses individual items without comprehensive discovery.

What output does backlog-scan produce?

Backlog-scan produces a comprehensive, numbered backlog with task titles, priorities, and actionable details formatted as task files, enabling systematic prioritization and execution planning.