skills.rest
majiayu000
majiayu000COMMUNITY
Explore all skills by this author →
GitHub

codex-retrospective

Review Codex usage history to propose AGENTS.md updates and tiny skills.

258|26|Updated Dec 9, 2025
Search Tags:#workflow#codex#retrospective#improvement#agents-md#tiny-skills
One-click install
npx skills add https://github.com/majiayu000/spellbook --skill codex-retrospective
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: codex-retrospective
Source: https://github.com/majiayu000/spellbook/tree/main/skills/codex-retrospective
Command: npx skills add https://github.com/majiayu000/spellbook --skill codex-retrospective

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Codex retrospectives help extract high-leverage improvements from usage history and convert them into minimal, reusable adjustments to AGENTS.md and tiny skills, enabling long-term fluency with your style and constraints.

Core Features & Use Cases

  • Structured prompts guide Codex to surface repeat friction, extract targeted changes, and maintain a lightweight, evidence-based upgrade loop.
  • Outputs include a Retrospective Summary, Proposed AGENTS.md Updates, and Proposed Tiny Skills to deploy next.
  • Works across weekly or monthly review cycles to accelerate learning and reduce re-explanation.

Quick Start

Run a focused retrospective on your Codex usage history to generate minimal, reusable improvements.

Frequently Asked Questions about codex-retrospective

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

FAQPage Schema
How do I automate Codex self-improvement loops from usage history?▼

You can automate Codex self-improvement loops by running a focused retrospective on recent usage history to identify high-leverage improvements and convert them into reusable adjustments for AGENTS.md and tiny skills.

How to perform a weekly or monthly retrospective for Codex agents?▼

Perform a weekly or monthly retrospective for Codex agents by applying structured prompts to surface repeat friction and extract targeted changes across your review cycles, maintaining a lightweight, evidence-based upgrade loop.

What outputs should a Codex retrospective generate for AGENTS.md updates?▼

A Codex retrospective generates outputs adhering to a strict structure including a Retrospective Summary, Proposed AGENTS.md Updates, and Proposed Tiny Skills to deploy next.

Can I use tiny skills to reduce re-explanation in my Codex workflow?▼

Yes, you can use tiny skills to reduce re-explanation by extracting minimal, reusable adjustments from your usage history, enabling long-term fluency with your style and constraints.

Does Codex retrospective work without external dependencies?▼

Yes, Codex retrospective works without external dependencies, requiring only your recent usage history to identify high-leverage improvements and generate proposed updates.

When should I run a retrospective to update AGENTS.md?▼

You should run a retrospective to update AGENTS.md during regular, scheduled weekly or monthly review cycles to accelerate learning and maintain an evidence-based upgrade loop.

Related Skills

Explore More Software Engineering Skills→
tygwantygwan
1

analytics

Aggregate and visualize CLI tool usage statistics from telemetry metrics.

Community
Intermediate
rstacruzrstacruz
13

reflect

Distill session learnings into AGENTS.md guideline improvements.

Community
Intermediate
CraftOS-devCraftOS-dev
375

heartbeat-processor

Process heartbeat triggers to execute scheduled tasks with tiered permissions.

Community
Advanced
alexknowshtmlalexknowshtml
137

reflect

Analyze Claude Code session transcripts to identify friction points and workflow inefficiencies.

Community
Intermediate