Hire an AI software launch evidence reviewer

Everything it can hand back

Your task sets the actual outputs. The specialist picks the methods that cover them.

• A dated decision log: test, revise or keep, with the evidence behind it
• An updated state file for your next run
• Missing facts, actions waiting for your approval and one next action

How it does the job well

• Start from the user problem and the evidence for it, then the solution.
• Write decisions down with the reason and the options you rejected.
• Keep scope small enough to ship and learn. Name what is out of scope.
• Define how success will be measured before building.

Mistakes it avoids

• Treating opinions or one loud request as proven demand.
• Specs that skip edge cases, error states or dependencies.
• Roadmaps with dates but no owners or confidence levels.

What a good result looks like

An example, not your task: You launched a SaaS product and cannot tell if you should change the positioning, find first customers or work on activation.

• agent-state.json. Product revision, buyer evidence, selected test, cash and time caps, the dates looked at, results seen and decision date.
• decision-log.md. The dated decision, the evidence behind it, the gaps that remain and one next action.
• Dated state. One label for this run: new, changed, unchanged, pending or blocked.
• Branch selected and reason. Which path was chosen (positioning, first customers or growth) and why.
• Missing facts and approvals. Facts still needed, actions waiting for your approval and the single next step.

How the work is checked

• Every requirement can be tested.
• Assumptions and risks are listed with how each will be checked.
• Dependencies and owners are named.
• States which checks passed, failed or could not run, and names missing inputs.

These checks have not been run against a model yet. Your AI uses them to check its own work.

A second opinion

Before it hands back, your AI reviews the draft the way Al Ries and Michael Porter would. This is a way of reviewing the work. It does not mean they wrote or endorse it.

Capabilities

• Compare saved state. Check which sources, dates or results changed since the last run.
• Choose the needed work. If the buyer or problem is unclear, work on positioning first. If positioning holds but there are no launch results, prepare a first-customer plan. If a self-serve funnel has mature results, assess activation and expansion. Results are mature when at least one full trial or buying cycle has passed since launch and you have signups, activations and paid customers for the same dates. If results are missing, too early or unchanged, return pending or unchanged instead of claiming growth. Keep assumptions separate from real customer outcomes. The branches run in order (positioning, then first customers, then growth). Only one runs each time.
• Save the decision. Record evidence, remaining gaps and one next action. Keep external action status explicit.

How the role works

1. Checks your task, context and available tools.
2. Picks the methods needed and prepares a work plan.
3. Carries out supported tasks, with each method's limits and review gates.
4. Checks the real deliverables and corrects supported faults.
5. Hands back files, evidence and unresolved gaps.

Your specialist's working setup

Download the base specialist without a task form or account. README.md and SETUP.md explain how to use the folder. SOUL.md defines its scope. INTAKE.md asks missing questions inside your AI. PLAYBOOK.md, SKILLS.md, BEST_PRACTICES.md, CHECKS.md and HANDOFF.md guide the work and its checked deliverables.

Complete associated skills, supporting files and licences are included. Choose only the methods needed for the actual task. Keep context, working files and accepted feedback in your AI project.

Use this specialist through MCP or the API

Call prepare_specialist with role_id saas-launch-review-agent only. No task is required to load it. Your AI asks what job you need done, what you want back and which relevant context it can use.

For task-led hiring, call find_specialist, explain the suggested specialists and let the user accept or change the team.

Complete ZIP: Download base specialist. API: GET /api/v1/specialists/saas-launch-review-agent/pack. The host AI supplies its own model, tools and execution; a pack does not connect accounts or run a task.

Inputs and helpful context

The specialist starts from the task brief. Only active methods require their inputs.

For the primary method, when used

• Your product description and who you think the buyer is
• Your launch plan, with your cash and time limits
• Your own targets, such as the signup-to-activation rate you need (the agent does not make targets up)
• Dated results, such as signups, activations and paid customers (if any)
• Your saved state file from the last run (if any)

Tools and access

• Any good chat AI: It plans from your notes. No code or files are needed.
• Claude, ChatGPT or Gemini: A chat AI can run the maths if you paste a table of channel results.
• Your product analytics export: Real sign-up and retention counts give better calls than guesses.

Functions and deliverables

• Software launch evidence reviewer: Review buyer evidence and launch results. Choose the next positioning, first-customer or growth check without inventing outcomes.
  • A dated decision log: test, revise or keep, with the evidence behind it
  • An updated state file for your next run
  • Missing facts, actions waiting for your approval and one next action
• SaaS market position adviser: Find a market position that sets your SaaS product apart. Give it your product and rivals. Get a clear position, buyer reasons and tests.
  • A check of how clear and different your message is today
  • A market position built in three layers
  • Headline and sub-headline options to test
  • A check of how easy the position is for rivals to copy
  • A step-by-step plan to test the position in small stages
  Source: GitHub Copilot community.
• First customer launch planner: Describe your new product. Get early buyer targets, a launch sequence and tests for reaching first customers.
  • A check of where your launch is stuck, if it has launched
  • A table of buyer targets with channel and reason
  • A week-by-week launch sequence
  • Two-week tests with success criteria and stop rules
  • Signals to watch, with your own targets
  • A list of missing facts
  Source: GitHub Copilot community.
• Self-service product growth planner: Share your product funnel. Get activation and expansion changes, with tests to see whether self serve growth fits.
  • A self-serve or sales-led fit check, with reasons
  • A time-to-first-value audit with fixes
  • A growth equation for each channel
  • A keep, scale or kill call for each channel
  • A three-case forecast and a one-page playbook
  Source: GitHub Copilot community.
• Product context file writer: Share product facts, audience and positioning. Get a reviewed context file that other marketing skills can reference.
  • A product marketing context file
  • A summary of what the file already holds
  • A change log of updates
  • Gaps you still need to fill
  Source: Corey Haines.
• Al Ries: Clarify what a brand should stand for using Al Ries’s recorded approach. Examine its category, focus, name and launch story.
  • Your problem worked on the way Al Ries would: their questions, their method, their push-back
  • The answer or the work, in Al Ries's voice, using their ideas by name
  • Anything that goes beyond their public record, marked
  • One suggested next step
• Michael Porter: Review a business strategy using Michael Porter’s recorded tools. Compare industry forces, choose a clear position and check which activities support it.
  • Your problem worked on the way Michael Porter would: their questions, their method, their push-back
  • The answer or the work, in Michael Porter's voice, using their ideas by name
  • Anything that goes beyond their public record, marked
  • One suggested next step

Tasks you can use it for

• You launched a SaaS product and cannot tell if you should change the positioning, find first customers or work on activation.
• New launch results came in and you need a dated decision on what to test, revise or keep.
• You want each review to start from the last one, with the evidence and targets saved.

Limits to check before starting

• You have no product or buyer yet: write down who the buyer is and what problem you solve first, or talk to five likely buyers.
• You only want to review one number, such as signups: check it in your analytics tool or a spreadsheet.
• You want messages sent, spending approved or accounts changed for you: do those yourself after you review the draft.

Best way to use this role

Give it one clear outcome, the actual files and a definition of done. Keep the work in one AI project so it can reuse checked context and earlier files.

• Open Claude, ChatGPT or Gemini and add the full skill pack. Use an AI that can read and save files.
• Share your product description, who you think the buyer is, and your launch plan. Add any new results with dates (signups, activations, paid customers). Add your saved state file if you have one.
• Tell it your cash and time limits, and the dates to look at.
• Ask it to follow the steps and mark any missing facts instead of guessing.
• Read the decision log. Approve any send, spend or account change yourself before it happens.

Read it as data

Read the complete role setup as JSON. Connect your AI to the AI Blocks MCP library and read aiblocks://roles/saas-launch-review-agent for its role instructions and pinned sources.

This is a prepared AI role. No task has run. Your AI carries out authorized work using its available tools and reports the actual deliverables and checks.