Hire an AI software release verification coordinator

Everything it can hand back

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

• A go, no-go or blocked verdict with the reasons
• The next version number and a changelog
• A deploy plan with a check after each step
• A rollback, with anything that cannot be undone said first
• A check that the live site serves the new build

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

• Setting a version before the first review is clear.
• Running the gate again and again with no change. Stop at the limit.
• Treating "the tests pass" as proof the release is safe.
• Writing a rollback that has never been read against the actual migration.
• Deploying because the person said "looks good" about the changelog. A yes must name the deploy.
• Claiming code works without running it or its tests.

What a good result looks like

An example, not your task: You ship software and want the same checks before every release.

• Verdict. A go, no-go or blocked at review, with the reasons.
• Version and changelog. The next version number, why it was chosen, and the changelog text.
• Blockers. Each blocker, where it is, what clears it, and whether it came back from a past release.
• Deploy plan. The steps in order, with a check after each step to see it worked.
• Rollback. The steps back, how long they take, and anything that cannot be undone.
• After deploy. Whether the live site serves the new build, and a pointer to the rollback if it does not.

How the work is checked

• Nothing was committed, tagged, deployed or migrated without a clear yes.
• The gate was run after the version was set, and the number of runs is stated.
• The rollback names what cannot be undone.
• Past blockers were checked.
• 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 its memory. Opens the last release and the blockers found in earlier releases, and lists what changed since.
• Review the deploy risks. A read-only check of data changes, settings, order of steps and rollback. A blocker stops the run before any version is set.
• Prepare the release. Works out the next version and writes the changelog. Both stay as local changes for you.
• Gate it, with a cap. Gives a go or no-go. After a no-go and your fixes, it runs again, up to three times in all.
• Plan the deploy and the way back. Writes the steps and the checks after each, with a longer plan when there is a data or settings change.
• Check after you deploy. Proves the live site serves the new build. If not, it points to the rollback and asks you.

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 release-manager 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/release-manager/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 project, with its history since the last release
• How the last release was done, the first time
• Known risks such as a data change or a partner that must update first (optional)
• The live link, once you have deployed

Helpful context (optional)

• The steps you followed for the last release
• A list of data or settings changes in this release

Tools and access

• Claude Code: It can read your commit history, keep the release folder and prepare the version and changelog.
• GitHub: It reads the commits and checks since the last release. You tag and publish.
• Claude or ChatGPT: Any good chat AI can write the plan from your description.
• Your host's dashboard: It shows the real deploy and rollback options, so you can check the plan against them.

Functions and deliverables

• Software release verification coordinator: Prepare a release and rollback plan, then check the actual live build. Keep blockers and deployment approval separate.
  • A go, no-go or blocked verdict with the reasons
  • The next version number and a changelog
  • A deploy plan with a check after each step
  • A rollback, with anything that cannot be undone said first
  • A check that the live site serves the new build
• Release readiness reviewer: Share a release or diff. Get a read-only report of migration, config, order and rollback risks to check.
  • Ranked list of blockers and risks
  • Evidence and owner for each item
  • Recommended fix for each item
  • Questions to confirm before release
  Source: sickn33.
• Software release preparer: Share your project. Review a version and changelog proposal, then confirm file updates and a release tag.
  • The version number updated in your version files
  • A changelog entry grouped into Features, Fixes and Documentation
  • A release commit and an annotated tag
  • A GitHub Release page, if you ask for one
  • A dry-run preview of every change
  Source: Jim Liu.
• Code release readiness checker: Run code checks before release. Get pass, fail and manual items, with blockers grouped by severity.
  • A pass, fail, skip or manual result for each check
  • A list of blockers with file locations
  • A go or no-go verdict
  • A prompt to re-scan after changes
  Source: Alireza Rezvani.
• Release safety planner: Get a safe deploy plan for your app: what to check first, the steps in order, how to roll back and how to know the release worked.
  • The deploy method that fits your host
  • A pre-release checklist in four groups
  • A step-by-step plan in five phases
  • Rollback triggers and steps for your host
  • Checks to run in the first hour and the next day
  Source: sickn33.
• Release readiness and live version checker: Check common blockers before a release. Then compare the live version with the one you meant to ship.
  • Pre-flight checklist results
  • Silent-failure checks
  • Proof the live version is new
  • Clear ship or hold verdict
  Source: sickn33.
• Live software release reviewer: Share a deployed service. Check code, build results, database behavior and live routes before calling the release ready.
  • Pass, fail or blocked verdict
  • Evidence for each gate
  • Steps to repeat each check
  • List of gaps that block release
  Source: sickn33.
• 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

• You ship software and want the same checks before every release.
• A release has a data change or a setting change and you want the rollback planned first.
• You want proof the live site runs the new version, not a hope.

Limits to check before starting

• You ship a static site or a one-off change with no build, deploy or rollback: use a plain release checklist instead.
• You want the agent to commit, tag, deploy or roll back for you: do those steps yourself, as this agent only plans and checks.
• You have no code history since the last release: ask a teammate for the list of changes first, or the changelog will be incomplete.

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.

• Start a chat with the agent. In Claude Code, add it as a skill or save it in .claude/agents, then tell it you are preparing a release.
• Read the verdict. Fix any blocker, then tell it so and it checks again.
• Read the plan and the rollback. Commit, tag and deploy yourself, or say yes to each step one by one.
• After you deploy, give it the live link and ask it to check.
• It saves the release for next time, including the rollback steps that worked.

Read it as data

Read the complete role setup as JSON. Connect your AI to the AI Blocks MCP library and read aiblocks://roles/release-manager 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.