QA Engineer for Manual Testing

Responsibilities

• Create and execute plans for manually testing and validating the Storyblok CLI.
• Reproduce and investigate bug reports.
• Run Storyblok CLI commands against a dedicated QA space to ensure features work as expected.
• Validate the DX of CLI features.
• Provide thoughtful feedback regarding incomplete, broken, or missing features.

When to use

Use this skill when the user requests manual testing of specific features and functionality.

Workflow

Preparation

1. Run pnpm nx build storyblok to build the latest version of the CLI.
2. Log in with the correct user: source ./.env.qa ./dist/index.mjs logout && ./dist/index.mjs login --token "$STORYBLOK_TOKEN" --region eu.
3. Create a comprehensive test plan based on the assignment.
   • Test all happy paths.
   • Test error scenarios.
   • Consider possible edge cases.

Performing tests

• Clean up the local and remote environment before and after every assignment by running ./.claude/skills/qa-engineer-manual/scripts/cleanup.sh. Don't hesitate to clean up between tests to start with a clean slate.
• Run CLI commands with ./dist/index.mjs (for available commands, run ./dist/index.mjs --help).
• Review command documentation in ./src/commands/COMMAND/README.md or ./src/commands/COMMAND/ACTION/README.md.
• Verify changes in the local file system or the Storyblok space.
• When using two spaces (for example one source and one target space), use STORYBLOK_SPACE_ID_TARGET for the target.
• Many commands require files in ./.storyblok/$STORYBLOK_SPACE_ID/COMMAND_DIR to perform a test. Create files in this directory when necessary. For example, when pushing a story, create it in ./.storyblok/$STORYBLOK_SPACE_ID/stories/SLUG_FAKE_UUID.json.

Interact with the Storyblok API

The Storyblok Management API provides access for verifying the state in Storyblok.

source ./.env.qa

# Get stories (paginated)
curl "https://mapi.storyblok.com/v1/spaces/$STORYBLOK_SPACE_ID/stories/" \
  -H "Authorization: $STORYBLOK_TOKEN"

# Get assets (paginated)
curl "https://mapi.storyblok.com/v1/spaces/$STORYBLOK_SPACE_ID/assets/" \
  -H "Authorization: $STORYBLOK_TOKEN"

Refer to Working with Assets when creating assets during manual testing.

Helper scripts

These scripts provide fast setup, inspection, and cleanup during manual tests. Each script loads .env.qa automatically. Paths are relative to this SKILL.md. For example, ./scripts/cleanup-remote.sh resolves to ./.claude/skills/qa-engineer-manual/scripts/cleanup-remote.sh.

Scenario seeds

Use these scripts to quickly seed QA spaces with predictable data for manual testing. Scenarios expect a clean space, so run cleanup first when you want a fresh start.

• Each scenario defaults to STORYBLOK_SPACE_ID unless you override it with --space.
• Paths are relative to this SKILL.md. For example, ./scenarios/has-stories.sh resolves to ./.claude/skills/qa-engineer-manual/scenarios/has-stories.sh.

Troubleshooting

• Load the STORYBLOK_SPACE_ID, STORYBLOK_SPACE_ID_TARGET, STORYBLOK_ASSET_TOKEN, and STORYBLOK_ASSET_TOKEN_TARGET environment variables by running source ./.env.qa (do not attempt to read this file).
• Save all temporary files in ./.claude/tmp.
• Find reports and logs in ./.storyblok/logs and ./.storyblok/reports.

Output

The final output must:

1. Provide thoughtful feedback regarding incomplete, broken, or missing features.
2. List all findings.

Review Template

Use the following template for the manual test review output:

# [Feature] Test Results

- **Feature/Command:** [for example, `storyblok push stories`]
- **Outcome:** [✅ PASS / ❌ FAIL / ⚠️ PARTIAL]
- **QA Space ID:** `$STORYBLOK_SPACE_ID`

## Test execution

- [ ] [Brief description of step 1].
- [ ] [Brief description of step 2].

## Findings & observations

- **Issue or bug:** [Description of what failed or behaved unexpectedly].
- **DX observation:** [Feedback on the developer experience and command output].

## Recommendations

- [Suggestions for improvements or missing features].