Prompts

A prompt is a guided, fill-in-the-blank request that the Cycle MCP Server provides ready-made. Instead of composing the perfect instruction yourself, you pick a prompt, answer one question in plain words, and the server expands your answer into a complete, Cycle-aware request for the agent. That request already tells the agent to read the right resources, explore the live app instead of guessing, and hand authoring off to a specialized agent.

Prompts are a practical on-ramp for Cycle MCP. They are especially useful if you are newer to working with AI agents: a short, vague request that would normally get an ineffective result becomes a well-structured workflow because the prompt supplies the structure for you.

How prompts are surfaced

Prompts are part of the MCP standard, but each agent host presents them differently:

Agent	How to invoke a prompt
Claude Code	Prompts appear as slash commands (type / and look for the cycle server's prompts), or in the prompt/command picker.
Cursor	Surfaced in the chat's prompt/command picker for the cycle MCP server.
GitHub Copilot in VS Code	Available through Copilot's MCP prompt surface in agent mode; support varies by Copilot version.

If your host does not list any Cycle prompts, confirm the server is connected and that you have signed in. Hosts that do not support MCP prompts at all will not show them. You can still describe the same request in plain language and the agent will use the Cycle tools and resources directly.

The agent also keeps a catalog of the available prompts at the cycle://prompts resource, so you can always ask: "What Cycle prompts are available?"

Available prompts

author-a-test

Use this when you want to create a new Cycle test but are not sure what details the agent needs from you.

• You provide: One input, request: in plain words, what the test should do. Put it all in one description: what you want to verify, the app or site under test (a URL or a name), the key steps or expected outcome if you know them, and any test data (logins, sample records, search terms).
• The agent then: Restates its understanding back to you for confirmation, asks for anything essential that is missing (especially the URL or screen to start from), explores the running application a step at a time so it never guesses at locators, hands the confirmed workflow to the feature-file-author subagent to produce a validated .feature file, and keeps a running record of the whole session in a MCP Prompt Logs/ folder at the root of your Cycle project.

You review the resulting .feature file and the log, then iterate.

debug-a-failed-run

Use this when a Cycle test has failed and you want help finding and fixing the cause.

• You provide: One input, request: in plain words, which test or .feature failed, what you observed (an error message, or the step that failed), and where the run output lives (the output directory or results .db) if you know it.
• The agent then: Gathers the concrete failure evidence first (it reads the structured cycle://run-errors/{output_dir} resource rather than guessing from the test name), reasons about the most likely cause (locator drift, timing, data, or authentication), inspects the live page if it looks like a locator problem, and only then proposes a fix with the evidence that points to it. It again keeps a running log under MCP Prompt Logs/.

begin-discovery

Use this to start or resume Discovery, the guided flow that turns your documentation and questionnaire answers into a recommended Cycle test plan.

• You provide: One optional input, system: the system under test. Leave it blank and the agent asks before doing anything else. Blue Yonder is the only supported system in this iteration.
• The agent then: Confirms the system, determines which Cycle project to work in (using an existing one, claiming the one it finds, or creating a new one), presents the input-source hub, drives whichever input source you pick, and returns you to the hub after each. When the questionnaire is complete and you ask for it, the agent generates the test plan.

This is the single entry point for Discovery. It replaced a standalone document-ingest prompt, so document ingest is now reached from the hub rather than invoked on its own.

Full walkthrough

The Discovery section covers this flow page by page, from the first invocation through generating and reviewing the test plan.

generate-tests-from-specs

Use this when you have a set of test specifications and want runnable Cycle tests from them. Today those specifications come from a Discovery document ingest run.

• You provide: The spec bundle root (required: the Test-Specs folder), the target test directory (required: where generated tests are written), and optionally the specific specifications to convert. Leave the last one blank to browse and choose interactively.
• The agent then: Lists and triages the bundle, flags likely duplicates and specifications an existing test already covers, takes your confirmed selection, and generates one test per specification, delegating the authoring to the feature-file-author subagent. Re-runs skip work that is already covered.

blueprint_onboarding

Use this the first time you set up test data with Blueprints, or when you want a guided path from an empty library to a working Blueprint.

• You provide: One optional input, system: the system you want to create data in, such as "Blue Yonder WMS", a REST API, or a SQL database. Leave it blank and the agent asks before doing anything else.
• The agent then: Confirms the system and which backend it speaks, reads the Blueprint setup, authoring, and schema resources rather than recalling the format, then reads your library's catalog and system configuration to learn the conventions already in place. From there it drafts a Blueprint, validates it structurally, runs it, and tears the data down.

Note that this prompt's name uses underscores while the others use hyphens. In Claude Code it appears as /cycle:blueprint_onboarding.

create-project

Use this to set up a new Cycle project from scratch.

• You provide: A project name, the parent directory to create it in, and your Cycle username.
• The agent then: Scaffolds the project directory tree, writes the .cycproj file and a user settings file, and confirms the paths it created.

You do not need this before Discovery. The begin-discovery prompt creates a project for you when there is not one already.