Hire an AI compliance evidence and request tracker

Everything it can hand back

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

• A gap list that shows what changed since the last review
• One draft request for each evidence owner, with what is needed and by when
• The path it took (rebuild the checklist, or check refreshed evidence) and why
• Lines marked MISSING for anything it could not confirm
• A short record to paste into your next review

How it does the job well

• Name the asset, the threat and the impact before recommending controls.
• Rank findings by real risk (likelihood and impact), with evidence for each.
• Prefer fixes that remove a problem over warnings that rely on people.
• Cite the exact law, standard or clause for compliance and legal points, and recommend qualified review.

Mistakes it avoids

• Repeating last review's findings as new ones.
• Treating an empty export as proof that nothing changed.
• Keeping private records in logs beyond the fields in the record.
• Calling installation a running service. Nothing runs unless the user starts it.
• Long lists of generic issues with no evidence they apply.
• Testing or scanning systems without permission.

What a good result looks like

An example, not your task: Your audit window opens in a few weeks and you need to know which controls still lack current evidence.

• What changed. Each control whose evidence expired, appeared or changed owner, with its old and new state. On a first run it says first baseline.
• Gap list. One row per control with control ID, evidence reference, checked date, owner, due date and a state of current, expired, missing or unverified.
• Draft requests for owners. One short request per owner naming the control, the proof needed, the date wanted and what acceptable proof looks like. Marked not sent.
• Path taken. Whether it rebuilt the checklist or only checked the refreshed evidence, and why the other path was skipped.
• MISSING lines and CHECKS MISSED. Every fact it could not confirm and every check that failed or could not run, or the word none.
• Record for next review. A small block to paste next time with control ID, evidence reference, scope, checked date, owner, due date and state.

How the work is checked

• Every control is graded, and each grade traces to a dated source or is labelled an assumption.
• No control is marked current on a closed task, a filled cell or an owner's say-so alone.
• Original files are unchanged, and any revised file has a new name or version.
• Both pieces ran, or the run says which one could not and why.
• 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

• Reads the last record. Uses the saved record if it matches this review, or sets a first baseline. The record holds control ID, evidence reference, scope, checked date, owner, due date and state.
• Checks what changed. Compares evidence dates, owners and requirement versions with the record. A missing export counts as unknown, not as no change. If nothing changed and nothing is due, it says so and stops.
• Picks the path. Changed requirements run the Prepare evidence for a compliance review workflow. Refreshed evidence alone runs the Spreadsheet checker on your evidence list.
• Grades each control. Marks each control current, expired, missing or unverified. Current means dated, in scope and tied to a named owner. A closed task alone never counts.
• Drafts the requests. Writes one request per owner that names the control, the proof needed, the date wanted and what acceptable proof looks like. Drafts only.
• Saves the review. Writes a run log, the step files and a new record, and keeps the old record. 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 compliance-evidence-request-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/compliance-evidence-request-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

• Your review scope, which is the framework, the controls in scope and the review dates (from your compliance lead or auditor)
• Your evidence list, one row per control with the file or link, its date, its owner and its due date
• Your last gap list and record (optional, needed to see what changed)
• Any requirement that changed since the last review (optional)
• Today's date and your time zone

Helpful context (optional)

• Your auditor's own list of evidence they will ask for.
• The name of the person who reviews proof and signs off.
• Last quarter's gap list.

Tools and access

• A chat AI that reads uploaded files: Claude, ChatGPT and Gemini can read your evidence list and the pack. In plain chat you save the record yourself.
• Your sheet or evidence tracker: Export it as a spreadsheet so each control has a date, an owner and a link on one row.
• A spreadsheet-capable AI: It can build and reopen a real evidence register. In plain chat you get the same register as text.

Functions and deliverables

• Compliance evidence and request tracker: Check new or expired evidence for a review. Keep gaps open until proof is checked and draft requests for the right owners.
  • A gap list that shows what changed since the last review
  • One draft request for each evidence owner, with what is needed and by when
  • The path it took (rebuild the checklist, or check refreshed evidence) and why
  • Lines marked MISSING for anything it could not confirm
  • A short record to paste into your next review
• Compliance evidence and gap organizer: Share your review scope, supplied framework and existing records. Get an evidence index, gaps and a preparation plan with owners.
  • Control questions, each with the proof it needs
  • An evidence register: control, file, date, scope and status
  • A gap list: what is missing, out of date or out of scope
  • A preparation plan with an owner, due date and dependency for each gap
  • A readiness note and any MISSING lines
• Spreadsheet checker: Check a spreadsheet for odd totals, broken formulas and messy data. Your original stays safe, and you get findings with proof and a new file.
  • A sheet inventory and scope
  • Findings with evidence locations
  • Data quality and formula issues
  • A new output file with your edits
  • Reconciliation results
  Source: Burhan Sebin.
• AI law gap checker: Share the AI rule and company practices. Check scope, compare requirements and plan gaps for legal review with owners and due dates.
  • A yes or no on whether the rule applies to you
  • The requirements pulled out of the rule
  • Each gap sorted into must do, should do, already met or accepted
  • A dated fix plan with owners and deadlines, saved as a file
  Source: Anthropic.
• 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

Tasks you can use it for

• Your audit window opens in a few weeks and you need to know which controls still lack current evidence.
• An owner says "I uploaded it" and you need to see if the file is dated, in scope and from the right system.
• A rule or framework requirement changed since your last review.
• Your evidence list lives in a sheet and last quarter's gap list is buried in a folder.

Limits to check before starting

• You have no evidence list or checklist yet. Start with the Compliance readiness checklist builder, or run the Compliance evidence and gap organizer workflow.
• You only want a spreadsheet's formulas or data checked. Use the Spreadsheet checker.
• You want a signed opinion that you comply, or legal advice. This reports readiness only. Your auditor or lawyer decides.
• You want the requests emailed for you. It drafts them 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 to your AI with the two pieces it uses, the Prepare evidence for a compliance review workflow and the Spreadsheet checker. If your AI cannot open them, paste both in with the agent.
• Say if this is the first run. If not, paste last time's record and gap list. Then attach your evidence list and review scope.
• Read "What changed" first, then the gap list. Open two or three evidence files yourself and check their dates.
• Edit the draft requests and send them yourself. Put the owners' replies into your evidence list.
• Run it again when evidence expires, a requirement changes or a due date arrives. In plain chat it cannot save files, so keep the record it hands back.

Read it as data

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