reextract-metrics

Name: reextract-metrics
Availability: InStock
Author: Sourcegraph

Re-extract task metrics from completed benchmark runs after extraction fixes.

32|4|Updated Feb 1, 2026

Search Tags:#analysis #metrics #benchmark #extraction #reprocessing #cost correction

Ready to run

npx skills add https://github.com/sourcegraph/CodeScaleBench --skill reextract-metrics

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: reextract-metrics
Source: https://github.com/sourcegraph/CodeScaleBench/tree/main/skills/reextract-metrics
Command: npx skills add https://github.com/sourcegraph/CodeScaleBench --skill reextract-metrics

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill addresses the need to update or correct task metrics after changes to the extraction logic or schema, ensuring accurate performance analysis of benchmark runs.

Core Features & Use Cases

Batch Re-extraction: Recalculates task_metrics.json for multiple benchmark runs.
Bug Fix Application: Applies corrections to extraction bugs or schema changes retroactively.
Use Case: After fixing a bug in the cost calculation script, use this Skill to re-process all historical runs to get accurate cost metrics.

Quick Start

Run the reextract metrics skill to refresh all task metrics.

Frequently Asked Questions about reextract-metrics

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

FAQPage Schema

What is the best way to apply cost corrections to historical benchmark run data?▼

Batch re-extraction recalculates task_metrics.json for completed benchmark runs by applying corrections to extraction bugs or schema changes retroactively. It walks active task directories containing result.json and enriches them with updated metadata.

What is the best way to apply cost corrections to historical benchmark run data?▼

Cost corrections are applied retroactively by re-processing historical runs to recalculate task_metrics.json. The Skill enriches the data with metadata and reports significant cost corrections after verifying metric changes.

How do I refresh task metrics from run data after a schema change?▼

To refresh task metrics after a schema change, Python scripts walk active task directories, re-run extraction pipelines on result.json files, and verify metric changes to ensure accurate performance analysis.

Do I need Python scripts to re-extract metrics from result.json files?▼

Yes, Python scripts are required to walk directories, re-run extraction pipelines, and verify metric changes across active task directories containing result.json, enriching them with updated metadata.

How does batch re-extraction handle significant cost corrections in benchmark data?▼

Batch re-extraction recalculates task_metrics.json across active task directories and reports significant cost corrections, ensuring accurate performance analysis of benchmark runs after extraction bug fixes or schema changes.