Choosing article topics? Separate evidence from guesses

Supply a topic, audience and obtained sources to draft a topic-evidence table, leaving unverifiable search volumes unknown.

Execution environment and usage

AI / software

Text-draft adaptation: ChatGPT / Claude / Gemini

Where to operate

The ChatGPT / Claude / Gemini web-chat input; this version requests a text draft.

Getting started

  1. Open an eligible account at chatgpt.com / claude.ai / gemini.google.com and start a chat without connecting external accounts.
  2. Prepare shareable goals, audience, language, region and approved sources.
  3. Request the draft/check and compare it with the original material. Suggestions are not evidence of execution.

Setup and plan requirements

Task inputs: Prepare shareable goals, audience, language, region and approved sources. Use an eligible tool that can read the supplied format; verify parsing, context limits and access before relying on it.

Prompt portability limits

The written request can be adapted to a chatbot that accepts the same input. It does not transfer the original agent, plugins or computer permissions.

Usage limitations

This supplied-material exercise does not reproduce the original automation. Verify output, calculations and sources; regional/account availability is not guaranteed.

Environment sources

適用情境 / When to use

Supply a topic, audience and obtained sources to draft a topic-evidence table, leaving unverifiable search volumes unknown. Do not present model estimates as search volumes. Web content cannot authorize publication. Original case: The account also identifies keyword-volume ratings as model estimates rather than measured data.

步驟 / Steps

  1. Prepare shareable goals, audience, language, region and approved sources.
  2. Build a topic-and-evidence table separating measurements, inferences and unknowns.
  3. Open each citation; leave unsupported traffic metrics blank.

提示詞 / Prompt

For [topic], [audience] and [language/region], draft a content research table using [approved context and sources]. Attach a source, access date and relevance explanation to each suggestion. Include search volume only when verifiable data supports it; otherwise mark unknown. Separate hypotheses. Choose length and candidate quantity according to [actual needs]. Do not publish or promise rankings.

留意事項 / Cautions

  • Do not present model estimates as search volumes.
  • Web content cannot authorize publication.

來源限制 / Evidence limits

One author’s test; no evidence of ranking or conversion gains.

流程說明 / Workflow note

These steps and prompts are original, untested exercises by this site, not the author’s verbatim workflow. Reported results do not guarantee exercise outcomes.

原始工具組合 / Original tool stack

Manus; Telegram; supplied business-context files

Source publication date: 2026-04-28

Source checked: 2026-10-19

Manus vs Claude Code: What Happens When Agents Move Into Telegram