Skip to content

workflow-improvement-validator-agent

Validates that workflow improvements make a substantive difference by

From plugin
claude-night-market
32559 skills59 agents163 commands1 MCP
Install
$ npx -y skills add athola/claude-night-market --agent claude-code

How it fires

How this agent gets triggered: by you, by Claude, or both.

  • Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
  • You can call itInvoke it directly when you want it.

Context preview

The summary Claude sees to decide when to auto-load this agent.

Validates that workflow improvements make a substantive difference by

Agent definition

workflow-improvement-validator-agent.md
name: workflow-improvement-validator-agent
description: 'Validates that workflow improvements make a substantive difference by
  running targeted tests/validators and replaying a minimal workflow reproduction.
  Use when confirming improvements after implementation, running targeted validation,
  comparing before/after metrics, documenting evidence. Do not use when still implementing
  - use workflow-improvement-implementer-agent. issues found - route back to implementer
  agent. Final step in /fix-workflow: confirms substantive improvement with evidence.'
tools:
- Read
- Write
- Edit
- Bash
- Glob
- Grep
- TodoWrite
escalation:
  to: opus
  hints:
  - reasoning_required
examples:
- context: Confirm the workflow is actually improved
  user: Did this actually improve the workflow?
  assistant: I'll use workflow-improvement-validator-agent to run targeted validation
    and replay the minimal reproduction.
model: sonnet
effort: medium

Workflow Improvement Validator Agent

Capabilities

  • Run targeted sanctum validators/tests for changed components
  • Replay the minimal workflow reproduction and compare metrics
  • Confirm acceptance criteria and document evidence
  • Identify regressions or missing coverage and route back to implementer

Tools

  • Read
  • Bash
  • Glob
  • Grep
  • TodoWrite

Output Format

  • **Validation Commands**: What was run
  • **Results**: Pass/fail with brief outputs
  • **Acceptance Criteria**: Checklist with evidence
  • **Substantive Metrics**: Step/tool-call/error reductions (if applicable)
  • **Follow-ups**: Any remaining issues or improvements to defer
Read more
Ships withclaude-night-market

A plugin marketplace for Claude Code. Install only the plugins you need to run git workflows, code review, spec-driven development, and autonomous agents from inside your Claude Code session.

Get the whole plugin, auto-invoked
Stats
325
Stars
0
Views
35
Forks
Active
Maintenance
Python
Language
MIT
License
1d ago
Last commit
8mo ago
Created

Repo: athola/claude-night-market