Hire an AI backup restore rehearsal reviewer

Everything it can hand back

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

• A recovery evidence table, one row per system: target, last restore date, actual result, evidence reference, gap and status
• A changed rehearsal plan saying which systems are due for a test, and why
• Open fixes, each with an owner, a due date and what proves it is fixed
• A short postmortem for each failed or missed test
• MISSING lines for results nobody wrote down, and a record to start the next review from

How it does the job well

• Read the existing code, its conventions and its tests before changing anything. Match the project's patterns.
• Make the smallest change that fully solves the request. Keep unrelated refactors out.
• Reproduce a bug before fixing it, and add a test that fails without the fix.
• Treat input from users, files and networks as untrusted. Validate it at the boundary.

Mistakes it avoids

• Repeating old recommendations as new findings.
• Treating a green backup job or a confident message as a passed restore.
• Describing a plan as a test that happened.
• Describing installation as a running automation.
• Claiming code works without running it or its tests.
• Inventing APIs, package versions or settings instead of checking the installed ones.

What a good result looks like

An example, not your task: Your last restore test was months ago and you cannot say if it passed.

• Evidence table. One row per system with its target, last restore date, actual result, evidence reference, gap and status.
• Due for a test. The systems that need a new test, why each one is due and the proposed scope inside the scope you approved.
• Changed rehearsal plan. The updated plan from the Backup and recovery planner, with a safe rehearsal, stop points and rollback steps from the Incident runbook writer.
• Postmortems. For each failed or missed test, what happened, what was found and what to fix.
• Open fixes. Each fix with an owner, a due date and what proves it is done.
• MISSING lines and CHECKS MISSED. Every result nobody wrote down and every check that could not run.

How the work is checked

• Every Verified row has a dated restore result and an evidence reference.
• No row is called recoverable on a backup log alone.
• Every Failed or Missed target row has a postmortem and a fix with an owner or MISSING.
• Proposed tests stay inside the approved scope.
• 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.

Capabilities

• Read the last review. Loads the saved record: system ID, plan version, last restore date, result, evidence reference and next review date. A first run builds a baseline with you.
• Find what is due. Marks a system due when a test is overdue, its evidence is missing or the system changed since its last test.
• Grade each system. Gives each system one status: Verified, Failed, Missed target, Not tested or Unknown. A backup job log alone reads as backup ran, restore not tested.
• Update the rehearsal plan. The Build a backup recovery rehearsal workflow updates the plan with the Backup and recovery planner, the Incident runbook writer and the Executive action register keeper. It proposes tests only inside your approved scope.
• Review failed tests. The Incident postmortem writer reviews each failed or missed test: what happened, what was found and what to fix. Each fix gets an owner.
• Check and save. Checks that no restore result is invented. Saves the files, a run log and the new record, and keeps the old record.

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 recovery-rehearsal-evidence-assistant 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/recovery-rehearsal-evidence-assistant/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

• Recovery targets for each system, approved by its owner: how long it may be down and how much recent data can be lost (from your continuity plan)
• Restore test records: the date, what was restored, how long it took and the result (from test notes, tickets or logs)
• Changes to each system since its last test, such as a new host or backup tool (from your change log)
• The last recovery plan or rehearsal plan (optional on a first run)
• The test scope you approve: which systems may be tested, and where (needed only if a new test is proposed)
• Last review's record (optional, from the second run on)

Helpful context (optional)

• Backup tool reports or logs.
• Your change log or release notes.
• A list of your systems, with an owner for each.

Tools and access

• A capable chat AI with file support: Keeps the record between reviews. In a plain chat tool you paste the record in each time.
• Claude, ChatGPT or Gemini with file support: They can read your backup records as pasted text or files. None of them can run a restore for you.

Functions and deliverables

• Backup restore rehearsal reviewer: Check restore evidence against recovery targets. Get a revised test plan, missed targets and follow-up actions.
  • A recovery evidence table, one row per system: target, last restore date, actual result, evidence reference, gap and status
  • A changed rehearsal plan saying which systems are due for a test, and why
  • Open fixes, each with an owner, a due date and what proves it is fixed
  • A short postmortem for each failed or missed test
  • MISSING lines for results nobody wrote down, and a record to start the next review from
• Backup restore rehearsal planner: Map what needs to be restored, write a safe rehearsal checklist and track the evidence needed to check recovery.
  • A list of your systems ranked by how fast you need them back
  • Recovery targets for each rank, using your numbers
  • A rehearsal checklist with prerequisites, stop points, rollback steps and one row per system
  • An evidence log template to record what happened and where the proof is saved
  • An action list with an owner and a date for each task and gap
• Incident postmortem writer: Give it the facts of an outage. Get a blameless postmortem with timeline, causes and follow-up actions.
  • A blameless postmortem document
  • A timeline of events
  • The main cause and the contributing causes
  • Action items with owners and due dates
  • A 60-minute plan for the review meeting
  Source: sickn33.
• Backup and recovery planner: Describe systems and agreed loss and downtime limits. Get a backup plan, restore steps and drills, with results when actually run.
  • A list of your systems ranked by how much you need them back
  • Recovery targets for each rank
  • A backup design, or a review of what you have
  • A restore guide for each system
  • A drill plan and a schedule for the next drill
  Source: RampStack.
• AI project context organiser: Share your project and next task. Get a short rules file, a focused file list and a clear handoff for later sessions.
  • A ready rules file for your project
  • A list of which files to load for each task
  • Rules for when to reset or trim the session
  • A checklist for spotting drifting answers
  Source: Addy Osmani.

Tasks you can use it for

• Your last restore test was months ago and you cannot say if it passed.
• You moved to a new database, host or backup tool, and the old test no longer applies.
• A restore test ran, missed its target, and nobody wrote down the fix.
• An auditor, insurer or client asks you to show that you can recover.

Limits to check before starting

• You have no backups or no test history yet. Use the Backup restore rehearsal planner workflow to write the first plan.
• A system is down right now. Follow your incident runbook and call your technical lead first. Use the Incident postmortem writer afterwards.
• You want step-by-step instructions to restore a system. Use the Incident runbook writer.
• You need a formal sign-off for an auditor. This agent sorts your evidence, but a qualified person must approve it.

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.

• Add the agent to Claude, ChatGPT or Gemini with its helpers, the Build a backup recovery rehearsal workflow and the Incident postmortem writer. Or download the pack and attach it.
• Paste your recovery targets, restore test records and change log. Say the review date. From the second review, paste last review's record too.
• Read the evidence table first. Check each Verified row against the real test record.
• Fill in each MISSING line, especially test results that were never written down.
• Approve any new test scope yourself and run the test yourself. The agent runs nothing.
• In a chat-only tool, copy the record it gives you and paste it in next time. Installing sets up no schedule, so you start each run.

Read it as data

Read the complete role setup as JSON. Connect your AI to the AI Blocks MCP library and read aiblocks://roles/recovery-rehearsal-evidence-assistant 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.