plan-backlog

Score GitHub Issues with the RICE framework and write a ranked backlog to agent/state/backlog.json.

5|1|Updated Mar 29, 2026
One-click install
npx skills add https://github.com/mataeil/OODA-loop --skill plan-backlog-mataeil
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: plan-backlog
Source: https://github.com/mataeil/OODA-loop/tree/main/skills/plan-backlog
Command: npx skills add https://github.com/mataeil/OODA-loop --skill plan-backlog-mataeil

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Scores open GitHub Issues using the RICE framework and outputs a prioritized backlog to agent/state/backlog.json.

Core Features & Use Cases

  • Scoring: Applies Reach, Impact, Confidence, and Effort to rank issues for planning.
  • Deduplication: Groups near-duplicate issue titles and collapses them to a single representative score.
  • State management: Reads/writes backlog.json and tracks run counts, status, and top score to enable iterative workloads.
  • Guardrails: Enforces safety checks and halts when needed to prevent unintended actions.

Quick Start

Run the plan-backlog skill to score open GitHub issues and write the ranked backlog to agent/state/backlog.json.

Frequently Asked Questions about plan-backlog

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

FAQPage Schema
How do I prioritize GitHub issues using RICE scoring?

To prioritize GitHub issues using RICE scoring, the skill computes Reach, Impact, Confidence, and Effort for each issue based on labels and body content, then sorts them by the resulting RICE score to produce a ranked backlog.

What is the RICE framework for backlog prioritization?

The RICE framework for backlog prioritization evaluates items by calculating Reach, Impact, Confidence, and Effort to generate a single score, enabling data-driven ranking of GitHub issues rather than relying on manual estimation.

Can I deduplicate duplicate GitHub issues before scoring a backlog?

Yes, you can deduplicate GitHub issues before scoring because the skill groups near-duplicate issue titles and collapses them into a single representative score, ensuring your prioritized backlog remains clean and actionable.

How do I apply RICE scoring to an existing backlog state file?

To apply RICE scoring to an existing backlog state file, the skill reads input from agent/state/backlog.json, scores the issues, and writes the updated ranked backlog back to the same file for iterative workloads.

Does the RICE backlog scoring process include safety checks for automated runs?

Yes, the RICE backlog scoring process includes safety checks such as halt detection and read-only mode enforcement to prevent unintended actions during automated workflows and iterative planning runs.