skills.rest
matejformanek
matejformanekCOMMUNITY
Explore all skills by this author →
GitHub

pg-shadow-implement

Compare AI-generated PostgreSQL implementation plans against upstream patch ground truth.

Updated Jun 1, 2026
Search Tags:#postgresql#calibration#planner#internals#patch-review#shadow-implementation
One-click install
npx skills add https://github.com/matejformanek/postgres-claude --skill pg-shadow-implement
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: pg-shadow-implement
Source: https://github.com/matejformanek/postgres-claude/tree/main/.claude/skills/pg-shadow-implement
Command: npx skills add https://github.com/matejformanek/postgres-claude --skill pg-shadow-implement

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill addresses the challenge of maintaining high-fidelity planning and implementation accuracy in an AI agent by creating a rigorous calibration loop against real-world PostgreSQL development threads.

Core Features & Use Cases

  • Phase E Calibration: Systematically compares AI-generated implementation plans against actual upstream PostgreSQL patches to measure accuracy and identify knowledge gaps.
  • Methodology Enforcement: Automates the seven-step shadow-implementation procedure, including spec extraction, design-level planning, and gap analysis.
  • Use Case: When a new feature is proposed on the pgsql-hackers mailing list, use this skill to test if the AI's internal planner would arrive at the same design decisions as the community, surfacing specific skill or corpus improvements.

Quick Start

Invoke the pg-shadow-implement skill by providing the URL of a PostgreSQL hackers thread to initiate a Phase E shadow-implementation run.

Frequently Asked Questions about pg-shadow-implement

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

FAQPage Schema
How do I calibrate AI-generated PostgreSQL implementation plans against upstream patches?▼

To calibrate PostgreSQL implementation plans against upstream patches, use a shadow-implementation methodology that systematically compares AI-generated designs against actual community patch ground truth to identify knowledge gaps.

What is the shadow-implementation process for PostgreSQL internals development?▼

Shadow-implementation for PostgreSQL internals is a seven-step procedure enforcing spec extraction, design-level planning, and gap analysis to validate AI planning workflows against real-world community development threads.

How do I test if my AI planner arrives at the same design decisions as PostgreSQL community proposals?▼

Test your AI planner against PostgreSQL community proposals by providing a pgsql-hackers mailing list thread URL to initiate a calibration run that compares agent design decisions against actual upstream patch ground truth.

When should I use a shadow-implementation methodology for PostgreSQL patch review?▼

Use shadow-implementation for PostgreSQL patch review when a new feature is proposed on the pgsql-hackers mailing list and you need to rigorously validate AI-driven planning, implementation, and review workflows against upstream ground truth.

Does shadow-implementation require external dependencies to validate PostgreSQL planner accuracy?▼

Shadow-implementation requires no external dependencies, relying solely on a multi-phase design-to-implementation comparison protocol to catalog gaps between AI-generated plans and actual PostgreSQL community patches.

Related Skills

Explore More Software Engineering Skills→
matejformanekmatejformanek

patch-submission

Prepare and submit PostgreSQL patches to pgsql-hackers and CommitFest.

Community
Advanced
nicholasglazernicholasglazer
27

eval

Evaluate corpus retrieval quality with Hit@5, MRR, and nDCG metrics.

Community
Intermediate
MartinNevlahaMartinNevlaha
1

spec

Coordinates spec-driven software development workflows through planning, implementation, verification, and learning phases.

Community
Advanced
RightNow-AIRightNow-AI
18.1k

postgres-expert

Analyze PostgreSQL query plans and recommend indexing strategies.

Official
Advanced