What is the purpose of the /eidos:goodjob command?

The command is used to record moments of high-quality AI output, including the context and the specific reasons why the output was successful, to help the AI calibrate its future behavior.

How do 'goodjob' files become part of the project specification?

Once captured, files are marked as 'unprocessed'. You can use the /eidos:toeidos command to extract the patterns into formal claims, rules, or spec updates, moving them to a 'processed' status.

When should I use the goodjob skill?

Use it whenever the AI provides an insight, a framing, or a structural solution that you find particularly valuable and want the AI to remember for future tasks.

Where are the calibration records stored?

Records are stored as Markdown files within the 'memory/' directory of your project, following a specific naming convention that includes a timestamp and a descriptive claim.

Eidos Goodjob Calibration

Name: Eidos Goodjob Calibration
Author: agenticnotetaking

byagenticnotetaking

•

Productivity & Workflow

Captures and documents successful AI outputs and design decisions to calibrate future performance based on project-specific context.

The Goodjob skill is a core component of the Eidos Spec Driven Development framework, designed to turn transient AI 'wins' into permanent project knowledge. By capturing specific moments where Claude's output aligns perfectly with intent—whether through a structural insight, a well-chosen metaphor, or a smart design decision—it creates a structured record of excellence. This skill ensures that the nuances of what makes a solution 'good' are documented in the project's memory, allowing future sessions to calibrate against proven patterns and maintain consistent quality across the development lifecycle.

Key Features

01Captures specific quotes and contextual insights from successful AI interactions

02Integrates directly with the memory directory for persistent project learning

03Automatically generates structured Markdown records with standardized timestamps

04Supports a workflow-based status system from 'unprocessed' to 'processed'

05Facilitates pattern extraction for updating project rules and specifications

0636 GitHub stars

Use Cases

01Recording design decisions that hit the mark to ensure future UI components follow the same 'taste'

02Documenting a particularly effective architectural framing or code metaphor for a complex system

03Capturing structural insights that solved a difficult bug to prevent future regressions

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add agenticnotetaking/eidos goodjob

For use in Claude.ai and ChatGPT

Download Skill