skills.rest
Lovuwer
LovuwerCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

testing-agent

Runs goal-based tests with MCP tools and pytest fixtures to verify AI agents.

Updated Feb 3, 2026
Search Tags:#debugging#pytest#agent-testing#MCP-tools#testing-agent#goal-based-testing
One-click install
npx skills add https://github.com/Lovuwer/Stupid --skill testing-agent
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: testing-agent
Source: https://github.com/Lovuwer/Stupid/tree/main/.claude/skills/testing-agent
Command: npx skills add https://github.com/Lovuwer/Stupid --skill testing-agent

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps teams verify that AI agents meet defined goals by running goal-based tests and iterating on failures.

Core Features & Use Cases

  • Test generation workflow: Generates guidelines and templates for constraint and success-criteria tests, enabling consistent test authoring.
  • Execution & debugging loop: Supports running tests, collecting results, and debugging failures to drive agent improvements.
  • Workflow integration: Works with MCP tools and pytest fixtures to manage test lifecycles in mock or real-mode environments.

Quick Start

  1. Load your agent goal from the agent.py file and define success criteria and constraints.
  2. Generate constraint test guidelines with the appropriate MCP tool and write tests using the provided file_header.
  3. Generate success criteria test guidelines and write tests similarly.
  4. Run the full test suite with the run_tests MCP command and iterate based on results.

Frequently Asked Questions about testing-agent

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

FAQPage Schema
How do I write goal-based tests for AI agents using pytest?▼

To write goal-based tests for AI agents, load your agent goals from agent.py, define success criteria and constraints, generate test guidelines using MCP tools, and write tests using the provided file_header with pytest fixtures.

What is the best way to verify that AI agents meet defined success criteria?▼

Verifying that AI agents meet defined success criteria involves running goal-based tests, collecting results, and debugging failures to drive iterative agent improvements within your development workflow.

Can I run agent tests in both mock and real-mode environments?▼

Yes, you can run agent tests in mock or real-mode environments by using MCP tools and pytest fixtures to manage test lifecycles and handle credentials appropriately.

How do I debug failing AI agent tests in a development workflow?▼

Debugging failing AI agent tests requires running the full test suite with the run_tests MCP command, collecting results, and iterating on agent improvements based on the failures identified.

How do I generate constraint test guidelines for an agent system?▼

Generating constraint test guidelines involves loading your agent goal, defining constraints, and using the appropriate MCP tool to create templates that enable consistent test authoring for your agent system.

Do I need MCP tools to execute a pytest test suite for agent validation?▼

Yes, MCP tools are required to generate guidelines, write tests, manage credentials, and execute the full pytest test suite for validating AI agent goals in development workflows.

Related Skills

Explore More Software Engineering Skills→
Owl-ListenerOwl-Listener
2.0k

design-brief

Define and document a design brief with problem space, audience, constraints, and success criteria.

Community
Intermediate
franroafranroa

dispatching-parallel-agents

Dispatch parallel agents to investigate and resolve independent test failures.

Community
Intermediate
btabaskabtabaska
1

Generate Mock

Generate deterministic mocks for pytest, Vitest, and MSW test environments.

Community
Intermediate
faisalanjumfaisalanjum
3

test-re-agent-field

Test the agent field in Skill definitions for direct tool access.

Community
Basic