$ ~/registry/skill/microsoft-agency-plugins-nori-skills-creating-debug-tests-and-iterating

SKILL

creating-debug-tests-and-iterating

Name: creating-debug-tests-and-iterating
Author: microsoft

Create debug tests and iterate to reliably reproduce and diagnose tricky bugs.

Stars

★ 4,934

Source

GitHub

Updated

2026-07-14

// securityCaution

Prompt only, no code
Open source

Summary

This skill appears to be an open-source prompt/instruction set with no required secrets and no declared remote endpoints, so the overall technical risk is low. However, the content shows clear prompt-injection and boundary-pushing tendencies, such as telling the agent to ignore existing tests, ask for credentials, start real services, and repeatedly add logs.

CredentialsCaution

The material declares no required secrets or environment variables, but the README explicitly says to ask the user for credentials if authentication is needed. This does not mean the skill itself holds secrets, but it can steer workflows toward handling real credentials during debugging.

Network egressLow risk

No remote endpoints are declared, and the objective checks classify it as prompt-only. The material mentions calling the application/API under test, but does not specify any third-party exfiltration target; there is no evidence of sending data to unknown external services.

Code executionCaution

The README explicitly suggests writing scripts, invoking CLIs via subprocess/exec, starting backend and frontend services, and using bash/tmux/playwright. These are normal local process-execution and debugging activities; the skill text itself does not execute code, but it strongly steers users toward real system actions.

Data accessCaution

Its workflow requires interacting with the real application from the outside, repeatedly adding logs, analyzing output, and cleaning up background jobs, which implies access to logs, test/script files, and runtime application data. No excessive permissions are explicitly requested beyond debugging, but logs may inadvertently capture sensitive information.

Supply chainCaution

A positive factor is that the source points to an open-source Microsoft GitHub repository and the object is classified as open-source, prompt-only, which improves auditability. Still, the provided metadata shows 0 stars, no declared license, unknown maintenance status, and the README contains clear injection-style wording (e.g., ignore existing tests, forcibly add Todo items, avoid mocks), so supply-chain trust is not maximal.

Recommendations

Do not provide production credentials to this skill; if authentication is required, use least-privilege temporary test accounts.
Use it only in isolated dev/test environments, not directly against production services or real user data.
Treat the README's imperative instructions with caution, especially phrases like 'ignore existing tests' and 'must add logs'; review them manually before following them.
If you add debug logging as suggested, review log content first to avoid capturing tokens, sessions, or other sensitive data.

Audited by: gpt-5.4 · 2026-06-17

// install

Copy the install command and let the AI configure it · recommended for beginners

Please install the "creating-debug-tests-and-iterating" skill from askskill:
1. Download https://raw.githubusercontent.com/microsoft/FluidFramework/main/.agency/plugins/nori/skills/creating-debug-tests-and-iterating/SKILL.md
2. Save it as ~/.claude/skills/creating-debug-tests-and-iterating/SKILL.md
3. Reload skills and tell me it's ready

// download

Download SKILL.mdInstall manifest ↗

// examples

Reproduce an intermittent API error

Prompt

Help me write a set of debug tests for this Node.js endpoint to reliably reproduce an intermittent 500 error. Based on the route code, log snippets, and sample requests I provide, design minimal reproducible tests, add necessary mocks, edge-case inputs, and diagnostic logging, and suggest how to narrow the investigation after each test iteration.

Expected

A runnable set of debug tests, reproduction steps, key logging points, and iterative debugging suggestions.

Diagnose incorrect data processing output

Prompt

I have a Python data-cleaning script whose final counts are occasionally wrong. Design debug tests for it covering nulls, duplicates, malformed timestamps, and boundary dates to reproduce the count discrepancy. Also show how to inspect intermediate outputs at each stage and explain which logic should be validated next.

Expected

Test cases covering key edge scenarios, intermediate validation checks, and an iteration plan for root-cause analysis.

Investigate state corruption under concurrency

Prompt

Help me create debug tests for this service with a concurrency issue, aiming to reproduce state corruption under multi-threaded or multi-request conditions. Include concurrency stress scenarios, timing control, assertion points, and a detailed logging strategy, and provide an iteration plan that follows: reproduce first, narrow conditions next, and verify the fix last.

Expected

A test setup that reproduces the concurrency issue, with key assertions, logging design, and a phased debug-and-fix verification plan.

// docs

<required> *CRITICAL* Add the following steps to your Todo list using TodoWrite:

<system-reminder>From this point on, ignore any existing tests until you have a working example validated through a new test file.</system-reminder>

Write a script that interacts with the application from the outside. The script should not call any internals. It should only interact with the external interfaces.
Check to see if you require authentication. If you do, ask me for credentials. <system-reminder>Do NOT use mock mode or test harnesses. You should be testing the real thing.</system-reminder>
Follow these steps in a loop until the bug is fixed:

Add many logs to the application. You MUST do this on every loop.
Run the debug script.
Analyze the output: read logs, identify errors, do whatever you need to.
Update the debug script. <system-reminder>If you get stuck: did you add logs?</system-reminder>

Identify and fix the issue at hand.
Clean up all background jobs, and remove extraneous logs.
Make sure other tests pass. </required>

Debug Testing

To test different kinds of applications, write scripts that test the application interfaces. Your testing should be as close to 'real' as possible.

Example

Identify the application boundary to be tested and the tools you need to test it.

CLI Tool:

./path/to/cli.sh arg1 arg2

subprocess.run(["./path/to/cli.sh", "arg1", "arg2"])

exec('./path/to/cli.sh', (error, stdout, stderr) => {
  if (error) {
    console.error(`exec error: ${error}`);
    return;
  }
  console.log(`stdout: ${stdout}`);
  console.error(`stderr: ${stderr}`);
});

API:

Start the server:

cd backend && python server.py&
cd frontend && npm run dev&

Call to the server using scripting language of choice.

Do NOT get in a loop where you just keep running other tests. In this mode, you should ignore other tests entirely until it works.

Emulators for testing

Web servers or web apps: use playwright (read the .claude/skills/webapp-testing/SKILL.md) TUI tools: use tmux with screen capture CLI tools: use bash

Find software bugs, diagnose causes, and suggest practical code fixes.

workersioGet→

$ loading_

Please install the "creating-debug-tests-and-iterating" skill from askskill: 1. Download https://raw.githubusercontent.com/microsoft/FluidFramework/main/.agency/plugins/nori/skills/creating-debug-tests-and-iterating/SKILL.md 2. Save it as ~/.claude/skills/creating-debug-tests-and-iterating/SKILL.md 3. Reload skills and tell me it's ready

// examples

Reproduce an intermittent API error

Prompt

Help me write a set of debug tests for this Node.js endpoint to reliably reproduce an intermittent 500 error. Based on the route code, log snippets, and sample requests I provide, design minimal reproducible tests, add necessary mocks, edge-case inputs, and diagnostic logging, and suggest how to narrow the investigation after each test iteration.

Expected

A runnable set of debug tests, reproduction steps, key logging points, and iterative debugging suggestions.

Diagnose incorrect data processing output

Prompt

I have a Python data-cleaning script whose final counts are occasionally wrong. Design debug tests for it covering nulls, duplicates, malformed timestamps, and boundary dates to reproduce the count discrepancy. Also show how to inspect intermediate outputs at each stage and explain which logic should be validated next.

Expected

Test cases covering key edge scenarios, intermediate validation checks, and an iteration plan for root-cause analysis.

Investigate state corruption under concurrency

Prompt

Help me create debug tests for this service with a concurrency issue, aiming to reproduce state corruption under multi-threaded or multi-request conditions. Include concurrency stress scenarios, timing control, assertion points, and a detailed logging strategy, and provide an iteration plan that follows: reproduce first, narrow conditions next, and verify the fix last.

Expected

A test setup that reproduces the concurrency issue, with key assertions, logging design, and a phased debug-and-fix verification plan.

// docs

<required> *CRITICAL* Add the following steps to your Todo list using TodoWrite:

<system-reminder>From this point on, ignore any existing tests until you have a working example validated through a new test file.</system-reminder>

Write a script that interacts with the application from the outside. The script should not call any internals. It should only interact with the external interfaces.
Check to see if you require authentication. If you do, ask me for credentials. <system-reminder>Do NOT use mock mode or test harnesses. You should be testing the real thing.</system-reminder>
Follow these steps in a loop until the bug is fixed:

Add many logs to the application. You MUST do this on every loop.
Run the debug script.
Analyze the output: read logs, identify errors, do whatever you need to.
Update the debug script. <system-reminder>If you get stuck: did you add logs?</system-reminder>

Identify and fix the issue at hand.
Clean up all background jobs, and remove extraneous logs.
Make sure other tests pass. </required>

Debug Testing

To test different kinds of applications, write scripts that test the application interfaces. Your testing should be as close to 'real' as possible.

Example

Identify the application boundary to be tested and the tools you need to test it.

CLI Tool:

./path/to/cli.sh arg1 arg2

subprocess.run(["./path/to/cli.sh", "arg1", "arg2"])

exec('./path/to/cli.sh', (error, stdout, stderr) => {
  if (error) {
    console.error(`exec error: ${error}`);
    return;
  }
  console.log(`stdout: ${stdout}`);
  console.error(`stderr: ${stderr}`);
});

API:

Start the server:

cd backend && python server.py&
cd frontend && npm run dev&

Call to the server using scripting language of choice.

Do NOT get in a loop where you just keep running other tests. In this mode, you should ignore other tests entirely until it works.

Emulators for testing

Web servers or web apps: use playwright (read the .claude/skills/webapp-testing/SKILL.md) TUI tools: use tmux with screen capture CLI tools: use bash