Summarize, index, and sort your folders.

What the AI will do

• Inspect the contents of a chosen folder collection, summarize the files and build a linked index.
• Organize the files into a useful structure, then verify the moves and index links.

What you can provide

• The folders or collection to organize and access to their contents.
• Write access enables actual sorting; with read-only access, the AI can provide an index and proposed filing map.

Where you'll take part

• Explain a retrieval preference the records do not reveal and settle consequential filing choices.
• Review the proposed organization before the AI makes agreed changes; enable a specific access step if needed.

What you'll get

An organized collection with a verified linked index and summaries—or, without write access, a clearly labeled partial result containing the index and proposed filing map.

Activity context

A folder is easier to use when you can tell what is inside, which files belong together and where to find the one you need.

For the user

Point the AI to the folder collection you want organized. It will inspect the contents and related project context, then propose a structure. You will settle meaningful filing choices before it sorts files and checks the result. Read-only access still supports a useful index, but leaves the sorting unfinished.

Guidance for the AI

These are general instructions, not a description of this user's particular AI system, tools, access, or circumstances. Adapt the method to the user's goal and the capabilities you can actually use, while preserving the activity's essential sequence, review points, and intended result.

Start with the activity and the user's context

Read the complete activity and guide the user through it. Use the activity prompt yourself; the user does not need to paste it again. Keep the responsibilities of the user, the AI, and any colleague clear.

Begin with what the user has already supplied. Establish the intended result and resolve missing context through focused questions when needed. Work within the activity's stated scope, permissions, required confirmations, and stopping points, including any deliberate order for introducing information. Make progress with the available information and reasonable, clearly identified assumptions where the activity allows them.

Check the capabilities needed for this work

Inspect relevant available tools, connectors, and data. Use small, relevant, non-destructive checks to establish whether necessary access or a capability actually works. Naming a product does not establish access to it. Let this activity determine what needs checking; do not impose a universal preliminary questionnaire or a full system audit.

When useful access is missing:

• Explain what you cannot reach and how that material would help.
• Investigate practical setup, connection, export, or other access alternatives before asking the user to gather material or troubleshoot.
• If progress depends on access, request only the specific user action or permission needed to enable it or provide the material another way. Continue useful independent work while awaiting the response.
• Offer to walk the user through an export or upload. Welcome files, links, copied text, rough notes, or an interview as supported by your system and the activity, and use the best available route. If the missing access is optional, continue with what is available.

Ask focused questions and respect the answers

Ask questions only when necessary to complete the work well or satisfy the activity's required review points. Build on answers already supplied. Ask one question at a time, or a small group of closely related questions. Wait on work that depends on the answer, and continue useful work that does not. Explain unfamiliar language with examples where helpful.

Confirm answers or decisions you supply on the user's behalf before treating them as the user's final answers or decisions. This does not require separate approval of every sourced finding; follow the activity's specific review requirements.

Keep evidence and independent contributions distinct

Distinguish evidence, inference, and missing information. Check your work against the source material.

Use independent agents when your system actually supports delegation and independent investigation or review would improve the work:

• Give each agent a bounded brief and the context appropriate to its role.
• Keep discovery independent until synthesis where the activity requires it. Preserve any requirement for the user or colleagues to contribute before seeing other explanations.
• Follow the activity's stated fallback and any required confirmation when independent agents or fresh contexts are unavailable. Preserve reviewer isolation where required; describe shared-context or sequential passes accurately.

Produce, check, and revise the result

Do the substantive investigation and production. Carry the activity through to the most complete, usable result within its scope, rather than ending with a plan or handing the main work back to the user. When missing information does not prevent useful work, proceed with clearly identified assumptions where the activity permits them.

Check the result against the activity's completion criteria. At each substantive delivery, state what is complete and any material unfinished work, assumptions, helpful missing information, and remaining uncertainty. Explain how these affect what the user can use, and help the user revise the result.

Preserve the activity's stopping point. Do not turn a rehearsal, investigation, or test-design activity into execution that the user has not authorized.

Your AI prompt

Summarize, index and sort a folder collection so I can find and understand its contents. Do the file inspection and organization yourself where capabilities permit. Ask questions one at a time and wait before changes that depend on consequential unresolved filing choices.

1. Locate the collection and understand retrieval

Start with the folder link, local path or description I supplied. Inspect available file or storage tools using a relevant listing and a small file read. Establish which locations and file types you can inspect and whether the intended organization can be written there. A folder listing does not establish content access or permission to move its files.

Search related project documents, emails and chats, including relevant WhatsApp messages where accessible, using folder names, recurring filenames, authors and project terms. Look for what the collection concerns, the work it supports, how people refer to its files, and any existing filing convention or links people rely on. Keep this context search limited to organizing this collection.

State the proposed boundary and retrieval purpose: for example, finding material by project, client, stage or document type. Ask about a missing preference only when it would change the structure. If the requested collection is too large for one pass, propose bounded coverage that can actually be completed and maintain a record of the remainder.

Investigate supported access or export options before requesting a user-only action. If content is unreadable, explain which summaries cannot be made. If write access is unavailable, continue with a linked index and proposed filing map and label this as a partial result; do not imply sorting has occurred.

2. Inventory and inspect the contents

Create an inventory containing:

• Each item's stable identifier when available, current name and location.
• Type, size or extent, accessible link and read status.
• Relevant dates, distinguishing a document date or reporting period from its last-modified timestamp. Identify aliases, shortcuts and duplicate references so the same item is not counted repeatedly.

Read enough of each file to summarize its actual subject, purpose and useful contents. For long files, inspect them in bounded sections and record incomplete coverage. For formats you cannot parse, try available previews or conversions without replacing the source; otherwise mark the content unread and use only the metadata you can verify. Never claim a filename-derived guess is a content summary.

Record relationships such as source and derivative, attachment and covering memo, related project, revision series or supporting dataset. Identify apparent duplicates as candidates by content comparison, not name alone. Do not delete or collapse them as part of routine sorting. Keep unresolved authority questions visible rather than silently declaring the newest file the master.

For large collections, assign agents bounded subfolders or file batches with actual access, identifiers, required summary fields and completion criteria. Track inventory coverage and reconcile files referenced across batches. Use waves as needed. Without agents, inspect manageable batches or fresh sessions with source and coverage handoffs; disclose sequential checking rather than calling it independent review.

3. Design a structure and filing map

Use the inspection and retrieval context to propose a simple folder structure and consistent naming where useful. Write a file-by-file map from existing location to intended destination, retaining identifiers and relationships. Show a few representative examples so I can judge how I would find something later.

Explain material alternatives when the collection supports different useful structures, such as project-first versus document-type-first. Recommend one based on actual use. Ask me to resolve consequential ambiguous placements, naming changes or treatment of shared material, one question at a time. Wait before affected moves. If the structure is already specified or approved, follow it without seeking the same decision again.

Consider how the storage system handles existing links and permissions during moves. Inspect its actual behavior or relevant current documentation when needed. Preserve the authorized sharing scope and stop for a specific decision if an intended move would change who can access a file. Choose supported stable links for the index where available. Retain a change record with original and destination locations so the organization remains understandable and reversible where the system permits.

4. Sort and build the index

After the filing choices are resolved, create the agreed destinations and move or rename the files in bounded batches. Check each operation's result before advancing. Do not overwrite colliding names; inspect the collision and choose a resolved destination. Track successful, failed and deferred operations explicitly, and repair recoverable failures before continuing.

Build the index from actual post-operation locations. Include a concise description, subject or project, relevant dates, important relationships and a working link for each covered item. Add a short explanation of the organization and a useful entry point for common retrieval needs. Use a document, spreadsheet or native index suited to the available environment and my intended use.

With read-only access, produce the same content-based index using existing links, alongside the unexecuted filing map. If only part of the collection can be read, mark its coverage and do not claim a complete index.

5. Verify the result and deliver it

Reconcile the final inventory against the starting inventory and change record: each covered file must be accounted for, with no unintended overwrite or disappearance. Verify every index link through an actual supported open or resolution check. Distinguish a directly checked link from one whose destination cannot be tested in this environment; repair broken links where possible and record remaining exceptions.

Give a separate reviewer the actual index, final inventory, filing map and representative underlying files, without your organizational rationale. Ask it to check whether summaries match contents and whether common retrieval tasks work. State the scope of any sampled content review. Use fresh-session review or disclose shared-context limits, and fix material findings.

Deliver the index and a concise account of the completed organization, remaining unread files, link-check exceptions and deferred moves. Show a few real retrieval examples. Distinguish actual sorting from a proposed map and stop at the agreed collection.

Done when

The agreed collection is summarized, indexed and sorted, with file accounting and link checks recorded; if access prevented sorting or complete inspection, you have the usable partial index and exact remaining filing map.

These instructions were written in September 2026 using the frontier models available at the time. You may now have better tools or models, or restrictions that make some tasks harder. Your AI should adapt to the tools and access you actually have and find the best available way to help. Everything here is a starting suggestion for you and your AI, not a proven best practice for every setup. Feel free to rewrite any of it to better meet your needs.