Test Scribe

Generate gherkin or manual scenarios based on your JIRA ticket using Ollama or Gemini. Create tickets test with one button.

AI-powered test case generator for JIRA. Turn ticket descriptions into Gherkin or Manual test cases in one click. Writing test cases from JIRA tickets by hand is slow, repetitive, and inconsistent — especially under sprint pressure. TestScribe reads your ticket, thinks like a tester, and generates ready-to-use Gherkin (Cucumber) or Manual test cases in seconds. Works directly inside Chrome on any JIRA Cloud instance. WHAT TESTSCRIBE DOES - Opens on any JIRA ticket — extracts the title and description automatically. - Generates Cucumber (Gherkin) BDD scenarios: Feature / Scenario / Given / When / Then. - Generates Manual test cases: step tables with Action / Data / Expected Result columns. - Edit, copy, or push each test directly to JIRA as a linked task. - Import tests straight into Xray Cloud with one click. - Enforce Step Consistency: Normalize phrasing before committing. - Context Enrichment: Auto-scans and injects Epic/Story descriptions inside the prompt context. - Switch between AI backends without leaving the popup. STEP CONSISTENCY & MERGE PANEL (New in v1.1.0) Keep your test suite clean and avoid duplicate step definitions: • Fuzzy Step Matching: Automatically compares generated steps against your team's existing synced library using Jaccard similarity. High matches (>=80%) show up in green, and medium matches (>=50%) in yellow. • Step Autocomplete: Type inside the step editor textareas to see recommendations from your synced project step dictionary. • Best Match Recommendations: Dynamically ranks candidate Jira test issues by overall step similarity to auto-select the best target ticket to merge your new scenarios into. • Clickable Saved Ticket Logs: Success logs show clickable hyperlinks leading straight to the new Jira issue in a new tab. FOUR AI BACKENDS — YOUR CHOICE 1. Ollama (local, recommended) — Runs entirely on your machine (e.g. Qwen2.5-Coder), no data leaves your network, no API costs. 2. Google Gemini (cloud) — Fast cloud generation, uses your Gemini API key. 3. Groq Cloud (cloud) — Ultra-fast cloud inference utilizing Llama 3, DeepSeek, or Mixtral models. 4. Custom OpenAI (local/cloud) — Connect to LM Studio, local hosts, or custom API gateways (DeepSeek Cloud, OpenRouter) via the standard OpenAI REST specification. REQUIRED: FIX OLLAMA CORS (Ollama users only) Chrome extensions send requests with a special Origin header that Ollama blocks by default, returning a 403 error. Run the command for your OS once, then restart Ollama. macOS: launchctl setenv OLLAMA_ORIGINS "*" Then quit and restart Ollama from the menu bar app. Linux (systemd): sudo systemctl edit ollama.service Add inside the file: [Service] Environment="OLLAMA_ORIGINS=*" Then run: sudo systemctl daemon-reload && sudo systemctl restart ollama Windows (PowerShell): [System.Environment]::SetEnvironmentVariable("OLLAMA_ORIGINS", "*", "User") Then quit and restart Ollama from the system tray. SETUP: OLLAMA (LOCAL AI — RECOMMENDED) 1. Download and install Ollama from ollama.com. 2. Pull the recommended model: `ollama pull qwen2.5-coder` (Optimised for structured output — produces better Gherkin and test tables than general-purpose models). 3. Click the TestScribe icon > Settings. 4. Select Ollama (Local), set host URL (default: http://localhost:11434). 5. Click Refresh next to the Model dropdown, select qwen2.5-coder, then Save. SETUP: GEMINI (CLOUD AI) 1. Get a free API key at aistudio.google.com/app/apikey. 2. Click the TestScribe icon > Settings. 3. Select Gemini (Cloud), paste your API key, then Save. TestScribe automatically selects the best available Gemini Flash model. SETUP: GROQ (CLOUD AI) 1. Generate an API key on the Groq Console (console.groq.com/keys). 2. Click the TestScribe icon > Settings. 3. Select Groq (Cloud), paste your API key, click Refresh, choose your preferred model, then Save. SETUP: CUSTOM OPENAI 1. Click the TestScribe icon > Settings. 2. Select Custom OpenAI. 3. Enter your Host URL (e.g., http://localhost:1234/v1 for LM Studio). 4. Enter an API key if required by your custom server. 5. Click Refresh, select your model from the dropdown, then Save. SETUP: XRAY INTEGRATION (OPTIONAL) 1. In JIRA: go to Apps > Xray > API Keys, create a key pair, copy Client ID and Secret. 2. Click the TestScribe icon > Settings > Xray Integration. 3. Enter your credentials, click Test to verify, then Save. TROUBLESHOOTING • 403 Forbidden from Ollama — Set OLLAMA_ORIGINS=* (see CORS section above). • Ollama unreachable — Start Ollama; check the host URL in Settings matches your port. • No models found — Run: `ollama pull qwen2.5-coder` • Gemini API Key missing — Settings > Gemini tab > paste key > Save. • Popup opens but no JIRA data — Navigate to a ticket URL (/browse/PROJ-123) first. • Tests not appearing — Right-click the popup > Inspect > check Console for errors. SOURCE CODE & FULL DOCUMENTATION https://github.com/pdaszko/TestScribe https://testitright.de/en/ai-tools/test-scribe