How do I see my most recent analysis?

Use the /runs latest command to instantly view details, status, and outputs of the most recent pipeline execution.

Can I compare results from two different runs?

Yes, the /runs compare {id1} {id2} command allows you to see side-by-side differences in findings, metrics, and agent performance.

Where are the run files stored?

All runs are stored as self-contained directories within the working/runs/ path of your project, containing their own outputs and states.

How do I save disk space with this skill?

Use the /runs clean command to automatically identify and delete runs older than 30 days after your confirmation.

AI Pipeline Run Manager

Name: AI Pipeline Run Manager
Author: ai-analyst-lab

byai-analyst-lab

•

Analíticas y Monitorización

Manages and analyzes past data pipeline runs to track performance, compare results, and maintain project cleanliness.

The AI Pipeline Run Manager skill provides a comprehensive dashboard for Claude Code to oversee the execution history of the AI Analyst plugin. It allows users to browse a detailed list of past analytical executions, inspect specific run states, compare results between different datasets or timeframes, and perform maintenance by cleaning up old data. By extracting data from local pipeline state files, it gives developers and data analysts full visibility into agent performance, success rates, and output artifacts directly within the terminal.

Características Principales

01Deep-dive inspection of specific run outputs and validation grades

02Comprehensive listing of past runs with status and agent progress tracking

0319 GitHub stars

04Smart partial-match run identification for quick navigation

05Automated cleanup of runs older than 30 days to optimize storage

06Side-by-side comparison of two distinct pipeline runs for metric analysis

Casos de Uso

01Troubleshooting failed runs by inspecting specific agent step logs

02Auditing the success rate of complex multi-agent data analysis pipelines

03Comparing insights and metrics between different versions of a dataset

Características Principales

01Deep-dive inspection of specific run outputs and validation grades

02Comprehensive listing of past runs with status and agent progress tracking

0319 GitHub stars

04Smart partial-match run identification for quick navigation

05Automated cleanup of runs older than 30 days to optimize storage

06Side-by-side comparison of two distinct pipeline runs for metric analysis

Casos de Uso

01Troubleshooting failed runs by inspecting specific agent step logs

02Auditing the success rate of complex multi-agent data analysis pipelines

03Comparing insights and metrics between different versions of a dataset