File Source

No JIRA. No GitHub. No credentials.

Pass a local file as --id and swayambhu-qa reads your requirements directly from it — no --source flag needed.

Usage

Always wrap the file path in double quotes — especially if the filename contains spaces, dashes, or special characters (common with macOS filenames):

/qa-pipeline --id "./story.md" --tool playwright
/qa-pipeline --id "requirements/login-feature.docx" --tool playwright,restassured
/qa-pipeline --id "specs/user-story.pdf" --tool restassured

# Filenames with spaces or special characters — quotes are required
/qa-pipeline --id "Petstore — Browse and Manage Pets (UI + API).txt" --tool playwright,restassured

When --source is omitted, the pipeline treats --id as a file path.

Supported formats

Format	Extension	What you need
Markdown	.md	Nothing — works out of the box
Plain text	.txt, .text	Nothing — works out of the box
Word	.docx	npm install mammoth
Word (legacy)	.doc	npm install word-extractor
PDF	.pdf	npm install pdf-parse@1.1.1

Optional dependencies are loaded at runtime — install only what you need. If a required package is missing, the pipeline prints an actionable install command and stops.

Tip: If you have a .rtf file, open it in any text editor (TextEdit → Format → Make Plain Text) and save as .txt — plain text works perfectly and requires no extra setup. RTF is not a supported format.

How it works

The file is parsed using the same extraction logic as IMS sources — acceptance criteria, test URLs, API endpoints, credentials, and priority are all pulled from the text using structured headings and patterns.

If your file doesn't use structured headings (common with Word docs and PDFs), the pipeline falls back to the full text and infers test scenarios from it directly.

Writing requirements for the best results

For structured extraction, use these headings in your file:

# Login Feature

## Summary
Users can log in with valid credentials and are redirected to the dashboard.

## Acceptance Criteria
- [ ] Valid credentials redirect to /dashboard
- [ ] Invalid password shows "Incorrect credentials" error
- [ ] Locked account shows "Account locked" message

## Test URL
https://staging.myapp.com/login

## API Endpoint
POST /api/auth/login

## Credentials
| Role  | Username        | Password   |
|-------|-----------------|------------|
| Admin | admin@test.com  | Admin@123  |
| User  | user@test.com   | User@123   |

Unstructured plain prose also works — the pipeline reads the full text and extracts scenarios from it.

Differences from IMS mode

Behaviour	IMS mode	File mode
Requires --source	Yes	No
Requires credentials in .env	Yes (per source)	No
Comments back to original ticket	Yes	Skipped — no remote ticket
issueId for file naming	From ticket ID	From filename (without extension)
AC parsing	From structured ticket fields	From headings or full text

File naming and issueId

The filename (without extension) is sanitized and used as the issueId throughout the pipeline — for log files, report files, and test case IDs. Sanitization lowercases the name and replaces any non-alphanumeric characters (spaces, dashes, dots, parentheses, special chars) with hyphens.

"Petstore — Browse and Manage Pets (UI + API).txt"
→ issueId: "petstore-browse-and-manage-pets-ui-api"
→ reports/results-petstore-browse-and-manage-pets-ui-api.json
→ logs/pipeline-petstore-browse-and-manage-pets-ui-api-2026-05-27.log

This means filenames with spaces or special characters work safely — always pass them in double quotes.