# Validate Failing Tool Output

Developers need predictable failure output when testing agent tool handling. This skill prints stdout, stderr, and a simulated stack trace before exiting with an error.

## Install

```bash
npx skillstore add enactprotocol/test-failing-test
```

## Metadata

- Status: approved
- Slug: enactprotocol-test-failing-test
- Version: 1.0.0
- Author version: 1.0.0
- Skillstore revision: r1
- Version status: valid
- Tree hash: b3f3627422f2dd33428b5a454955fac50f6c513af4125919c95a8f3b031eb592
- Author: EnactProtocol
- GitHub username: EnactProtocol
- License: MIT
- Repository: https://github.com/EnactProtocol/enact/tree/main/test-tools/failing-test
- Ref: a06681402992ceae98ba04d54cfd4ab004862696
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: confirmation\_required
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Quality score: 38
- Quality tier: warning
- Public page: https://skillstore.pages.dev/skills/enactprotocol-test-failing-test
- Manifest: https://skillstore.pages.dev/api/skills/enactprotocol-test-failing-test/manifest

## Capabilities

- Accepts an optional message string for the failure run.
- Prints a start marker and the supplied message to stdout.
- Writes error markers and stderr text before exiting.
- Creates a simulated Error object and prints its stack trace.
- Exits with status code 1 for failure-handling tests.

## Use Cases

- Agent Error Handling Checks: Verify that Claude, Codex, or Claude Code displays tool stdout, stderr, and stack traces clearly.
- Pipeline Failure Visibility: Run a predictable failing command to confirm logs and nonzero exits appear in automation.
- Support Reproduction: Generate consistent failure output when documenting how an agent environment reports tool errors.

## Prompt Templates

### Run Default Failure

```
Run the failing test tool with its default message and summarize the visible stdout and stderr.
```

### Use Custom Message

```
Run the failing test tool with the message: Release check failure. Tell me which lines appear before the stack trace.
```

### Compare Output Channels

```
Run the tool and separate the stdout lines from the stderr lines in your response.
```

### Audit Failure Handling

```
Run the tool with a unique trace label, then explain whether the environment preserved the label, stack trace, and failed exit.
```

## Limitations

- It is designed to fail on every run.
- It does not run assertions or validate application behavior.
- It does not capture logs from other tools or services.
- Command argument handling depends on the skill runner environment.

## Best Practices

- Use short, non-sensitive messages because the message is printed to logs.
- Run it only in test environments where a nonzero exit is expected.
- Review stdout, stderr, and exit status separately when validating agent behavior.

## Anti Patterns

- Do not use it as a real health check because it always fails.
- Do not pass secrets or personal data in the message value.
- Do not rely on it to validate application test suites or business logic.

## Security Audit

- Audited at: 2026-07-06T15:31:14.13\+00:00
- Summary: Static analysis reported no findings, and the JavaScript file only prints output, creates a simulated error, and exits. Manual review found one semantic concern: the skill metadata places user input inside a shell-style command template.

## Stats

- Views: 150
- Downloads: 8
- Favorites: 0
- Popularity score: 0