audit-performance

Orchestrate multi-agent audits to identify bottlenecks and generate prioritized remediation plans.

1|Updated Feb 25, 2026
One-click install
npx skills add https://github.com/danalexilewis/taskgraph --skill audit-performance-danalexilewis
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: audit-performance
Source: https://github.com/danalexilewis/taskgraph/tree/main/.cursor/skills/audit-performance
Command: npx skills add https://github.com/danalexilewis/taskgraph --skill audit-performance-danalexilewis

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Streamlines complex performance auditing for a Dolt-backed task graph by orchestrating a lead architect and parallel sub-agents to identify bottlenecks, anti-patterns, and remediation priorities. It produces an actionable remediation plan, with explicit findings and tasks.

Core Features & Use Cases

  • Orchestrated multi-agent audits: coordinates a pre-compute agent plus five scanners to gather fast, structured insights.
  • Dolt-aware analysis: assesses Dolt-specific bottlenecks, diff costs, and branch strategies to guide improvements.
  • Remediation planning: ranks findings by severity and impact and outputs implementable tasks for engineers.
  • Use Case: when you need a prioritized plan to improve performance across schema, queries, and Dolt operations.

Quick Start

Invoke the audit-performance skill on a target area to generate a remediation plan.

Frequently Asked Questions about audit-performance

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

FAQPage Schema
How do I run a performance audit on a Dolt-backed task graph codebase?

Dolt performance audits identify bottlenecks, anti-patterns, and remediation priorities by orchestrating a lead architect with parallel sub-agents. This approach assesses Dolt-specific diff costs and branch strategies to produce an actionable, prioritized plan for engineers.

How does multi-agent performance analysis work for task graph codebases?

Multi-agent performance analysis coordinates a pre-compute agent and five scanners to gather structured insights. It reads shared context notes to ensure cross-scan consistency, then synthesizes findings into implementable tasks ranked by severity and impact.

Can I use this performance audit skill to assess Dolt-specific bottlenecks?

Yes, the performance audit skill is Dolt-aware and explicitly assesses Dolt-specific bottlenecks, diff costs, and branch strategies. It evaluates schema, queries, and Dolt operations to guide targeted improvements across the task graph.

What is the best way to prioritize performance remediation tasks for engineers?

The best way to prioritize remediation tasks is orchestrating parallel scanners that rank findings by severity and impact. This generates a structured remediation plan with explicit findings and implementable tasks tailored for engineers.

Do I need shared context notes to run a Dolt performance audit?

Shared context notes are not required but are highly recommended to enrich the performance audit analysis. The skill reads and references available context notes to ensure cross-scan consistency across the five parallel scanners.

When should I not use an orchestrated multi-agent approach for performance auditing?

An orchestrated multi-agent performance audit is not suited for lightweight or single-issue investigations where a full pre-compute phase and five scanners would be excessive. It is designed for comprehensive audits across schema, queries, and Dolt operations requiring a prioritized plan.