improvement-loop-planner

Aggregate project signals into measurable improvement loop proposals with success criteria.

2|2|Updated Mar 17, 2026
One-click install
npx skills add https://github.com/mathiasbourgoin/roster --skill improvement-loop-planner
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: improvement-loop-planner
Source: https://github.com/mathiasbourgoin/roster/tree/main/.opencode/skills/improvement-loop-planner
Command: npx skills add https://github.com/mathiasbourgoin/roster --skill improvement-loop-planner

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill eliminates the risk of vague, unbounded improvement work by turning scattered signals from project knowledge bases, code, tests, issues, and CI systems into structured, measurable improvement loop proposals with clear success criteria.

Core Features & Use Cases

  • Intent Alignment: Aggregates project goals and constraints from documentation and knowledge bases (if present) to ensure improvements align with team priorities.
  • Signal-Based Targeting: Inspects mechanical evidence like failing tests, CI failures, flaky tests, and issue backlogs to identify high-impact, low-risk improvement targets.
  • Structured Proposals: Outputs 1 to 5 prioritized loop proposals with explicit metrics, verification commands, keep/discard rules, and guardrails against circular completion checks.
  • Use Case: A team with recurring flaky tests and open issues with clear acceptance criteria can use this skill to generate prioritized, executable improvement plans ready to hand off to an execution skill.

Quick Start

Use the improvement-loop-planner skill to generate a set of bounded, measurable improvement loop proposals for your current project or a specific area you want to optimize.

Frequently Asked Questions about improvement-loop-planner

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

FAQPage Schema
How do I plan bounded improvement loops for software projects?

You can create measurable improvement loops by aggregating signals from knowledge bases, code, tests, issues, and CI systems. This skill outputs prioritized proposals with explicit metrics, verification commands, and guardrails against circular completion checks.

What is the best way to prioritize high-impact, low-risk improvements?

The best way to prioritize high-impact, low-risk improvements is to inspect mechanical evidence like failing tests, CI failures, and flaky tests, aligning those signals with project goals and constraints to generate structured proposals.

How do I avoid vague, unbounded refactoring work?

Avoid vague, unbounded refactoring work by generating structured improvement loop proposals that include explicit metrics, deterministic verification signals, keep/discard rules, and guardrails against circular completion checks.

Can I use CI analysis to generate executable improvement plans?

Yes, CI analysis provides mechanical evidence like CI failures and flaky tests that this skill aggregates to generate prioritized, executable improvement plans ready to hand off to an execution skill.

Does improvement planning work without a project knowledge base?

Improvement planning works by aggregating project goals and constraints from documentation and knowledge bases if present, relying primarily on mechanical evidence from code, tests, issues, and CI systems to identify targets.