Hire an AI employee feedback action reviewer

Everything it can hand back

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

• New themes since the last review, with small groups held back
• Each agreed action marked done with proof, in progress, overdue or blocked
• A two-minute progress brief for leaders
• A draft update to staff covering only what changed
• Changes to your response plan for you to approve
• A short record with no quotes and no names

How it does the job well

• Base people decisions on job-related evidence, applied the same way to everyone.
• Write in plain, respectful language the person affected could read.
• Keep personal data to what the task needs, and handle it privately.
• Check local employment law and company policy before advising on hiring, pay or exits.

Mistakes it avoids

• Showing a result that could point to one person.
• Treating a closed ticket or a manager's word as proof.
• Quoting a comment, even a short one, in the record or the update.
• Calling installation a running service. Nothing runs unless the user starts it.
• Biased wording in job ads, reviews or interview questions.
• Plans with no owner, timing or way to check progress.

What a good result looks like

An example, not your task: Your survey results came in and you need to turn them into actions with owners.

• Progress brief. Two or three plain lines for leaders on what moved since the last review and what is stuck. A first run says first baseline.
• New themes. Each new theme in plain words, marked recurring or one-off, with a proposed action and owner to approve. No quotes.
• Held back for group size. How many themes were held back because the group was below your minimum. It shows no detail that could point to a person.
• Action tracker. Each agreed action with owner, deadline and status of done with proof, in progress, overdue or blocked, plus the proof and its date.
• Draft update to staff. A short update that covers only what changed and promises nothing you have not agreed. Marked not sent.
• Record for next review. A small block with theme IDs, action IDs, owners, deadlines, proof status and last review date. No quotes and no names.

How the work is checked

• No theme, count, quote or team name from a group below the minimum appears anywhere.
• No quote and no name appears in the record or the saved files.
• No action is done without dated proof.
• The draft update promises nothing the user has not agreed.
• 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 Kim Scott would. This is a way of reviewing the work. It does not mean they wrote or endorse it.

Capabilities

• Reads the last record. Uses the saved record or sets a first baseline. The record holds theme IDs with no quotes, action IDs, owners, deadlines, proof status and the last review date.
• Checks the trigger. New anonymized feedback arrived, or an agreed action is overdue or blocked. If neither, it says so and stops.
• Applies the group-size rule. Any theme from fewer people than your minimum is held back. It shows no count, quote or team name that could point to a person.
• Finds new themes. For new feedback it runs the Turn employee feedback into owned actions workflow. Recurring themes are kept apart from one-off comments, then turned into risks, gaps in proof and proposed actions.
• Checks agreed actions. The Executive action register keeper reads your agreed action list and proof, plus the proposed actions from the step before, and compares proof with each commitment. An action moves to done only with dated proof. Overdue and blocked actions are listed.
• Drafts the staff update. Writes an update that covers only what changed, in plain words. It promises nothing you have not agreed.
• Saves the record. Writes a record with no quotes and no names, keeps the old one, and lists MISSING lines and CHECKS MISSED.

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 employee-feedback-action-review-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/employee-feedback-action-review-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

• The anonymized feedback, with no names or identifying details (an export or summary from your survey tool)
• How many people each team or theme covers
• The action list you agreed, with each action, owner, deadline and status
• Proof of progress since the last review, such as a published policy or a dated meeting note
• Your minimum group size for reporting (your own rule, from your HR policy)
• What staff were told about how their feedback would be used
• The last review's record (optional)

Helpful context (optional)

• Your HR policy on survey privacy and minimum group size.
• The "you said, we did" update from last time.
• Headcount for each team.

Tools and access

• A chat AI that reads uploaded files: Claude, ChatGPT and Gemini can read your feedback summary and the pack. In plain chat you save the record yourself.
• Your survey tool's export: Give your AI the group-level summary and strip names and free-text details first.
• A chat AI that reads files: It reads all the comments and the pack together. Nothing is sent or changed for you.

Functions and deliverables

• Employee feedback action reviewer: Track what changed after employee feedback. Get new themes, overdue actions and a draft progress update.
  • New themes since the last review, with small groups held back
  • Each agreed action marked done with proof, in progress, overdue or blocked
  • A two-minute progress brief for leaders
  • A draft update to staff covering only what changed
  • Changes to your response plan for you to approve
  • A short record with no quotes and no names
• Employee feedback action planner: Share staff feedback, survey scope and privacy limits. Get themes, proposed actions with owners and a clear staff-response draft.
  • A theme summary that shows what many people raised and what only a few did
  • A list of groups held back because they are too small to report
  • Risks and evidence gaps for the themes the feedback supports, with a proposed action for each
  • Priority actions, each with an owner, a check date and a sign it worked
  • A draft staff update, ready for your review before it is sent
• Action item tracker: Share meeting notes or email threads. Get action items with owners, deadlines and unresolved commitments.
  • action-register.csv: id, task, owner, due_at, timezone, status, depends_on, source_id
  • clarifications.md
  • proposed-task-changes.md
• Employee feedback survey planner: Plan staff surveys and focus groups, or review supplied results. Get questions, a reading plan and clear follow-up actions.
  • Survey questions grouped by theme, with an answer scale
  • A focus group or stay interview guide
  • A way to read the results and find the top themes
  • An action plan with owners and dates
  • A guide for managers to discuss team results
  • A launch and follow-up plan
  Source: Tuan Duc Tran.
• me.md builder: Answer a few questions, or let your AI use what it remembers. Get one private file that tells any AI who you are.
  • One me.md file with up to ten dated sections
  • A list of facts taken from memory, for you to check
  • A short note of what it removed for privacy
  • The two next steps, to save it and set it up
• Kim Scott: Prepare feedback using Kim Scott’s recorded approach. Ask for criticism, describe the work clearly and follow up with care.
  • Your problem worked on the way Kim Scott would: their questions, their method, their push-back
  • The answer or the work, in Kim Scott'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

• Your survey results came in and you need to turn them into actions with owners.
• Three months after you promised changes, staff ask what happened.
• An action you agreed is overdue or blocked and you want to see which, and why.
• You are preparing a short "you said, we did" update for staff.

Limits to check before starting

• A complaint, grievance or discipline case about a named person. Use your HR process. This reads group feedback only.
• A team too small to hide who said what. Report across several teams instead, or leave that group out.
• You want to design the survey itself. Use the Employee feedback survey planner.
• You want to prepare a one-to-one with someone you manage. Use the One-on-one agenda and feedback planner.
• You want the update sent for you. It drafts and stops.

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 with the pieces it uses, the Turn employee feedback into owned actions workflow and the Executive action register keeper. If your AI cannot open them, paste both in with the agent.
• Before you paste anything, check that your AI tool's data terms and your HR policy allow staff feedback to go into it. Staff feedback leaves your organisation when you paste it. Paste group-level summaries, not raw comments. Remove names and any detail that points to one person. Tell the AI your minimum group size. Say what staff were told about how feedback would be used.
• Paste the feedback, the action list and the proof. If this is a repeat, paste last time's record.
• Read "Held back" first. Then check each action marked done against its proof.
• Edit the staff update and send it yourself. In plain chat it cannot save files, so keep the record it hands back.
• Next, agree who approves the update before it goes out, file the approved record in your HR folder, and set the date of the next review.

Read it as data

Read the complete role setup as JSON. Connect your AI to the AI Blocks MCP library and read aiblocks://roles/employee-feedback-action-review-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.