Skip to content
CodexWork
Any AIChatGPTClaudeCodexCopilotGeminiGrokOther agents

Use materials in your tool

All tool notes →

Turn research into a decision memo

Hand over research and criteria. Get a decision memo that names the unknowns that matter.

04060809
Designed method · documented tool support

An internal local synthesis used the exported instruction and synthetic brief with dated notes from two primary pages observed on 23 October 2026. The original saved memo, source table, brief, instruction and dated notes are included in this kit; this establishes only that bounded research path. No ChatGPT Work or Claude Cowork account, file task, editable-output tool or unattended run was tested. Documentation and account conditions can change.

A source-backed recommendation is not a hands-on product test.

Read the full instruction ↓

What you supply

  • Named options and the decision to make
  • Decision criteria, deadline and authorized research scope

What it saves

  • Decision memo
  • Dated source table
  • Open questions and next check

Recorded offline synthesis · fictional decision context

Where to try a first supervised file task

A conditional decision, with dated sources

  • Compare ChatGPT Work and Claude Cowork against the actual task
  • Use an already verified environment rather than assume account access
  • Check editable output and two-file permission before choosing
  • Read the full memo and the two-source table
Read the complete example →Starting materials →Editable Markdown ↓

The instruction

Use my brief to compare the named options for the stated purpose. Confirm the decision criteria and constraints before filling any material gap. If browsing is available, check current primary sources and save a source table with title, URL, date checked and the specific claim supported. If browsing is unavailable, use only the supplied sources and say which claims remain unchecked.

Save decision-memo.md and sources.md. The memo should include the decision, relevant criteria, a compact comparison, a conditional recommendation, risks and the next action. Separate documented facts, reasonable inferences and your judgment. State unknown access, cost or operational constraints rather than guessing. Cite claims where they are used; do not fabricate citations or treat a documentation page as proof that my account can perform the task.

Recommend only as far as the evidence supports. If a decisive unknown would change the choice, state the question and the smallest check that could settle it. Do not buy, subscribe, connect accounts or contact anyone. Return the saved files and any limits on the research.

Show plain text
Use my brief to compare the named options for the stated purpose. Confirm the decision criteria and constraints before filling any material gap. If browsing is available, check current primary sources and save a source table with title, URL, date checked and the specific claim supported. If browsing is unavailable, use only the supplied sources and say which claims remain unchecked.

Save decision-memo.md and sources.md. The memo should include the decision, relevant criteria, a compact comparison, a conditional recommendation, risks and the next action. Separate documented facts, reasonable inferences and your judgment. State unknown access, cost or operational constraints rather than guessing. Cite claims where they are used; do not fabricate citations or treat a documentation page as proof that my account can perform the task.

Recommend only as far as the evidence supports. If a decisive unknown would change the choice, state the question and the smallest check that could settle it. Do not buy, subscribe, connect accounts or contact anyone. Return the saved files and any limits on the research.

Stops in this kit

04

Unknown access, cost or operational limits

06

Buying, subscribing or contacting anyone

08

Criteria and constraints, before gaps are filled

09

Decisive unknowns, each with its smallest check

Checks to run

  1. Each decision-changing factual claim has a working primary-source link.
  2. Facts, inference and recommendation can be distinguished.
  3. Unknown eligibility or operational constraints remain explicit.
  4. The next action settles a real uncertainty rather than just adding research.
Open the checklist →

What has been tested

Designed method · documented tool support

An internal local synthesis used the exported instruction and synthetic brief with dated notes from two primary pages observed on 23 October 2026. The original saved memo, source table, brief, instruction and dated notes are included in this kit; this establishes only that bounded research path. No ChatGPT Work or Claude Cowork account, file task, editable-output tool or unattended run was tested. Documentation and account conditions can change.

Checked 2026-10-23

Limits

  • A source-backed recommendation is not a hands-on product test.
  • Documentation, availability and prices can change; check before committing.
  • The agent needs actual browsing access to refresh sources.

Optional tool setup

Using a kit

  1. Give the agent a specific decision and a short list of options.
  2. Allow primary-source research, or supply sources and state their dates.
  3. Read the recommendation alongside the unresolved questions before acting.
Use a kit in your tool →

ChatGPT

  1. Paste the instruction, attach the sample files or your own, and read the stops first.
  2. Pick Work for a multi-step task and supply the context it needs.
  3. For reuse, follow the documented skill or plugin setup on your surface. Downloading a kit does not install it.

Check the permissions for the tools and actions you use. The kit's stops are instructions to the agent; they do not create an enforced approval setting.

ChatGPT · dated notes →

Claude

  1. Paste the instruction, attach the sample files or your own, and read the stops first.
  2. In Cowork, choose the authorized folder containing your inputs where that feature is available. A generic kit-to-skill import was not established by the reviewed pages.
  3. In Claude Code, use CLAUDE.md for project guidance. AGENTS.md is read by default on supported versions (2.1.277+) only when project-path CLAUDE.md or CLAUDE.local.md does not take precedence; session and configuration exceptions apply.

Check the controls in your Claude surface. Anthropic describes approval handbacks, but the kit's stops do not enforce a product permission boundary.

Claude · dated notes →

Codex

  1. Paste the instruction, attach the sample files or your own, and read the stops first.
  2. For project guidance, use AGENTS.md. For a reusable skill, follow the supported setup, such as .agents/skills/<name>/SKILL.md in a repository.
  3. Check the sandbox and approval settings before you start the job.

Codex has sandbox and approval settings. Check them before a job that touches real files.

Codex · dated notes →

Copilot

  1. Paste the plain instruction into a Copilot chat you can access, and read the stops first.
  2. Check which file inputs and actions your tenant's actual surface supports before supplying files.
  3. Keep the kit as a manual prompt. The reviewed posts do not document importing it as a skill, plugin or agent.

Ask your administrator which actions, permissions and approval controls your organization has enabled. A kit's stops do not create those controls.

Copilot · dated notes →

Gemini

  1. Paste the instruction, attach the sample files or your own, and read the stops first.
  2. Where Skills is available, copy or import the skill using that surface's documented setup.
  3. Gemini app and Workspace skills do not sync. Set up the skill separately on each available surface you use.

Some Gemini Tasks may require confirmation. Check current eligibility and permissions before real account work; the kit's stops are instructions, not enforced approval settings.

Gemini · dated notes →

Grok

  1. Paste the plain instruction as a message and read the stops first.
  2. If Bot is available to you, show it the workflow; xAI describes routine learning by following along.
  3. Keep the instruction as plain text; we could not confirm SKILL.md support.

xAI describes approval handbacks for Grok Bot. Check what your account and connected apps allow before real actions.

Grok · dated notes →

Other agents

  1. Paste the plain instruction into an agent that accepts text, then supply inputs through its supported route and read the stops first.
  2. If your agent reads AGENTS.md, save the instruction there.
  3. If the named client supports Agent Skills, use the documented skill folder location and setup for that client and version.

This varies by agent. Look up its approval settings; the kit's stops stay in the instruction wherever you paste it.

Other agents · dated notes →
Markdown · README.md
# Turn research into a decision memo

Hand over research and criteria. Get a decision memo that names the unknowns that matter.

## Start

1. Edit [brief.md](research-to-decision-memo/files/brief.md) and [input-template.md](research-to-decision-memo/files/input-template.md).
2. Supply authorized inputs to your own file-capable agent with [instruction.txt](research-to-decision-memo/files/instruction.txt).
3. Inspect the saved outputs using [checks.md](research-to-decision-memo/files/checks.md) before using them. Keep original inputs.

## Inputs

- Named options and the decision to make
- Decision criteria, deadline and authorized research scope

## Outputs

- Decision memo
- Dated source table
- Open questions and next check

## Steps

- Give the agent a specific decision and a short list of options.
- Allow primary-source research, or supply sources and state their dates.
- Read the recommendation alongside the unresolved questions before acting.

## Instruction

Use my brief to compare the named options for the stated purpose. Confirm the decision criteria and constraints before filling any material gap. If browsing is available, check current primary sources and save a source table with title, URL, date checked and the specific claim supported. If browsing is unavailable, use only the supplied sources and say which claims remain unchecked.

Save decision-memo.md and sources.md. The memo should include the decision, relevant criteria, a compact comparison, a conditional recommendation, risks and the next action. Separate documented facts, reasonable inferences and your judgment. State unknown access, cost or operational constraints rather than guessing. Cite claims where they are used; do not fabricate citations or treat a documentation page as proof that my account can perform the task.

Recommend only as far as the evidence supports. If a decisive unknown would change the choice, state the question and the smallest check that could settle it. Do not buy, subscribe, connect accounts or contact anyone. Return the saved files and any limits on the research.

## Decisions that need your context

- Criteria and constraints, before gaps are filled
- Decisive unknowns, each with its smallest check
- Unknown access, cost or operational limits
- Buying, subscribing or contacting anyone

## Settled by the brief

No separately annotated settled rules are supplied. Follow the instruction as written.

## Check the output

- Each decision-changing factual claim has a working primary-source link.
- Facts, inference and recommendation can be distinguished.
- Unknown eligibility or operational constraints remain explicit.
- The next action settles a real uncertainty rather than just adding research.

## Limits

- A source-backed recommendation is not a hands-on product test.
- Documentation, availability and prices can change; check before committing.
- The agent needs actual browsing access to refresh sources.

## Materials

- [examples/research-to-decision-memo/designed-example.md](research-to-decision-memo/files/examples/research-to-decision-memo/designed-example.md)
- [examples/research-to-decision-memo/input.md](research-to-decision-memo/files/examples/research-to-decision-memo/input.md)
- [examples/research-to-decision-memo/offline-synthesis/decision-memo.md](research-to-decision-memo/files/examples/research-to-decision-memo/offline-synthesis/decision-memo.md)
- [examples/research-to-decision-memo/offline-synthesis/input.md](research-to-decision-memo/files/examples/research-to-decision-memo/offline-synthesis/input.md)
- [examples/research-to-decision-memo/offline-synthesis/instruction.txt](research-to-decision-memo/files/examples/research-to-decision-memo/offline-synthesis/instruction.txt)
- [examples/research-to-decision-memo/offline-synthesis/source-notes.md](research-to-decision-memo/files/examples/research-to-decision-memo/offline-synthesis/source-notes.md)
- [examples/research-to-decision-memo/offline-synthesis/sources.md](research-to-decision-memo/files/examples/research-to-decision-memo/offline-synthesis/sources.md)
- [examples/research-to-decision-memo/source-notes.md](research-to-decision-memo/files/examples/research-to-decision-memo/source-notes.md)

## Evidence

[Recorded context and file identities](research-to-decision-memo/files/evidence.json)

Designed method · documented tool support

An internal local synthesis used the exported instruction and synthetic brief with dated notes from two primary pages observed on 23 October 2026. The original saved memo, source table, brief, instruction and dated notes are included in this kit; this establishes only that bounded research path. No ChatGPT Work or Claude Cowork account, file task, editable-output tool or unattended run was tested. Documentation and account conditions can change.

Observed or source-checked: 2026-10-23.

## Revision and corrections

Revision: 2.1.0. Lifecycle: active.

[Current kit](https://codexwork.com/kits/research-to-decision-memo/) · [Changes and corrections](https://codexwork.com/changes/)

[Rights and notices](research-to-decision-memo/files/LICENSES.md) · [Sources](research-to-decision-memo/files/sources.md) · [Manifest](research-to-decision-memo/files/manifest.json)

Downloading these files does not run an agent, install a skill, create a schedule, or grant file or account permissions.
Open Markdown file ↗
SKILL.md
---
name: research-to-decision-memo
description: "Hand over research and criteria. Get a decision memo that names the unknowns that matter."
---

# Turn research into a decision memo

Use my brief to compare the named options for the stated purpose. Confirm the decision criteria and constraints before filling any material gap. If browsing is available, check current primary sources and save a source table with title, URL, date checked and the specific claim supported. If browsing is unavailable, use only the supplied sources and say which claims remain unchecked.

Save decision-memo.md and sources.md. The memo should include the decision, relevant criteria, a compact comparison, a conditional recommendation, risks and the next action. Separate documented facts, reasonable inferences and your judgment. State unknown access, cost or operational constraints rather than guessing. Cite claims where they are used; do not fabricate citations or treat a documentation page as proof that my account can perform the task.

Recommend only as far as the evidence supports. If a decisive unknown would change the choice, state the question and the smallest check that could settle it. Do not buy, subscribe, connect accounts or contact anyone. Return the saved files and any limits on the research.

## Required inputs

- Named options and the decision to make
- Decision criteria, deadline and authorized research scope

## Expected outputs

- Decision memo
- Dated source table
- Open questions and next check

## Checks

- Each decision-changing factual claim has a working primary-source link.
- Facts, inference and recommendation can be distinguished.
- Unknown eligibility or operational constraints remain explicit.
- The next action settles a real uncertainty rather than just adding research.

## Limits

- A source-backed recommendation is not a hands-on product test.
- Documentation, availability and prices can change; check before committing.
- The agent needs actual browsing access to refresh sources.

Revision: 2.1.0. [Read the complete kit](files/README.md) before use. This optional format is not a claim of support by every tool.
Open SKILL.md ↗