retro

Conducts a retrospective on a coding session to suggest agent environment improvements.

Updated Jun 28, 2026
One-click install
npx skills add https://github.com/AO-HyS/aohys.com --skill retro-ao-hys
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: retro
Source: https://github.com/AO-HyS/aohys.com/tree/main/.agents/skills/retro
Command: npx skills add https://github.com/AO-HyS/aohys.com --skill retro-ao-hys

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? After a coding session, it is hard to know why an AI coding agent struggled, made mistakes, or wasted tool calls. This Skill reviews session logs and proposes concrete improvements to the agent's environment so future runs go better. ## Core Features & Use Cases - Session log analysis: Reads primary sources for a specified session (defaulting to the current one) to find friction points. - Categorized improvement candidates: Evaluates navigation, automated checks, coding standards, AGENTS.md size, tool economy, no-op instructions, and information access. - Prioritized recommendations: Presents improvement candidates to the user ordered by severity. - Use Case: After a long debugging session where the agent repeatedly failed to find a config file, run a retrospective to add a navigation pointer to AGENTS.md and a lint check that would have caught the error. ## Quick Start Ask the agent to run a retrospective on the current coding session and suggest environment improvements.

Frequently Asked Questions about retro

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

FAQPage Schema
How do I run a retrospective on an AI coding session?▼

Invoke the retro skill and optionally specify which session to review; it defaults to the current session. It reads the session logs, identifies improvement candidates across categories like navigation and automated checks, and presents them ordered by severity.

What kinds of improvements does a coding session retrospective suggest?▼

It suggests navigation pointers for hard-to-find files, automated checks like linting or tests to catch mistakes, new or clarified coding standards for the reviewer agent, trimming of oversized AGENTS.md files, and fixes for token-inefficient tool calls.

Should coding standards be enforced by the implementation or review agent?▼

The review agent should enforce coding standards. The implementation agent has the most context pressure from exploration, coding, and debugging, while the review agent receives only a diff and has spare capacity for standards enforcement.

When should instructions go in AGENTS.md versus CODING_STANDARDS.md?▼

AGENTS.md is pushed into every agent's context, so it should be used sparingly, mostly for navigation pointers. CODING_STANDARDS.md is read only during review, making it the right place for enforcement rules, with doc pointers if it exceeds 1,000 lines.

What are no-op instructions in agent steering files?▼

No-ops are instructions in steering files that do not actually change the agent's behavior. The retrospective flags them so large, unwieldy steering files can be trimmed to only effective guidance.