Test Bruno

Use this skill when working with Bruno API tests. Bruno stores collections as plain-text .bru files in the repo, so tests are reviewed and versioned like code.

Requires

• test-automation-guidelines
• test-api

Apply the shared API principles from test-api first; the rules below are Bruno-specific additions.

Workflow

1. Confirm the collection layout and environments (bruno.json, environments/).
2. Refer only to the reference file relevant to the task:
   • Collection and environment structure: references/collections.md
   • Assertions, scripting, and running: references/scripting-and-assertions.md
3. Apply the principles and templates from that file.

Bruno Rules

Files and structure

• Keep .bru files in version control and review them like code — never treat the collection as a throwaway export.
• One request per .bru file, named for the behaviour it verifies.
• Organise requests into folders by resource/feature (folders are plain directories on disk).

Variables and environments

• Reference the base URL and other config as variables ({{baseUrl}}), defined per environment under environments/.
• Store secrets as secret variables (.env / secret vars), never committed literals. Commit an example env with placeholders only.
• Use {{process.env.VAR}} or Bruno secret vars for credentials so they stay out of the committed .bru.

Assertions

• Prefer the declarative assert block for status and simple field checks.
• Use the tests block with expect for schema/shape and richer logic.
• Assert status and key fields; avoid asserting volatile values (see test-api assertions guidance).

Scripting

• Use pre-request scripts to obtain tokens or set up data, and post-response scripts for extraction/cleanup.
• Capture only the specific values later requests need (e.g. bru.setVar('orderId', res.body.id)); avoid leaking broad state.
• Keep scripts small and deterministic; do not add arbitrary waits.

Anti-Patterns

• Committing .bru files or environment files that contain real tokens or API keys.
• Hardcoded full URLs instead of {{baseUrl}}.
• Bloated tests scripts re-implementing assertions that assert handles declaratively.
• One .bru file exercising many endpoints.
• Relying on collection-runner order for tests that should be independent.