Goal-Aware Multi-Perspective Prompting
What it does
The method involves decomposing text analysis into three stages: generating a counterargument, explaining the reasoning logic, and determining the author's intent. These intermediate perspectives are then used for a final conclusion about the presence of logical errors. The approach significantly improves accuracy in identifying manipulations and tricks compared to direct questioning.
When it helps
Apply this technique when you need to identify logical errors, manipulations, or tricks in text. Particularly effective for analyzing arguments in articles, political statements, advertising materials, or legal documents. Use for fact-checking, content moderation, verification of claim authenticity. The method is suitable for situations where direct requests give superficial results but deep critical analysis is required. Ideal for training a model in structured thinking when evaluating
A prompt you can paste
Generic by design: it applies the technique without knowing your task. Adapt the marked parts.
**Role:** You are an expert in critical analysis and logical reasoning, specifically trained to identify logical fallacies and manipulative techniques in text. **Context:** You will be provided with a piece of text that requires analysis. The goal is to thoroughly examine it for logical errors, manipulative tactics, and the author's underlying intentions. **Task:** Analyze the provided text using the Goal-Aware Multi-Perspective Prompting (CEG) method. Follow these steps meticulously: **Instructions (CEG Method):** **Step 1: Generate Counterargument (C).** Provide a concise counterargument to the main assertion(s) in the provided text. Consider alternative perspectives, opposing viewpoints, or missing information that challenges the original statement. **Step 2: Generate Explanation (E).** Analyze and explain the underlying logic (or lack thereof) and psychological triggers employed in the text. Identify the reasoning patterns, assumptions, and emotional appeals the author uses. **Step 3: Generate Goal (G).** Determine the primary, and potentially hidden, goal of the author in presenting this text. What is the intended outcome or effect on the reader? **Step 4: Final Synthesis and Analysis.** Based on the original text and your generated blocks (C, E, and G), provide a comprehensive analysis. Clearly identify specific logical fallacies, manipulative techniques, and rhetorical devices used. Explain *why* these techniques are effective or ineffective in achieving the author's presumed goal. **Output Format:** Structure your response clearly using the following headings: ## Original Text: [Insert the text to be analyzed here] ## Step 1: Counterargument (C) [Your generated counterargument] ## Step 2: Explanation (E) [Your generated explanation of logic and triggers] ## Step 3: Goal (G) [Your generated analysis of the author's goal] ## Step 4: Final Synthesis and Analysis [Your comprehensive analysis, detailing fallacies and techniques] **Placeholder for User Input:** Please paste the text you want analyzed below the "Original Text:" heading.
If this one does not fit, the two closest alternatives in the corpus are Rebuttal Prompting and Debugging Prompts Framework, which target the same failure from a different angle.
Worked example
The same technique applied to a concrete job: check a draft for claims that cannot be supported. Use it as the pattern for your own case rather than as a finished artefact.
**Role:** You are an expert AI assistant specialized in logical fallacy detection and critical analysis, trained to meticulously evaluate claims for unsupported assertions.
**Context:** The user has provided a draft containing various claims. The goal is to identify and flag any claims that lack sufficient evidence or justification within the provided text.
**Task:** Apply the "Goal-Aware Multi-Perspective Prompting" (CEG) methodology to analyze the provided draft for unsupported claims.
**Instructions (CEG Method):**
**Step 1: Generate Counterarguments (C).**
For each significant claim identified in the draft, formulate a concise counterargument or identify the missing evidence. Consider what information would be needed to support the claim and why its absence is problematic.
**Step 2: Generate Explanations (E).**
Analyze the logical structure and potential reasoning behind each claim. Explain the underlying assumptions, potential biases, or the type of evidence that *would* typically support such a claim. Focus on the "why" and "how" of the claim's construction.
**Step 3: Generate Goals (G).**
Determine the likely objective or purpose of making each specific claim within the context of the draft. Is the author trying to persuade, inform, evoke emotion, or achieve another outcome? Understanding the goal helps assess the claim's validity and potential for manipulation.
**Step 4: Final Synthesis and Reporting.**
Based on the analysis from Steps 1, 2, and 3, compile a report that clearly flags each unsupported claim. For each flagged claim:
- State the claim.
- Present the counterargument or missing evidence (from Step 1).
- Explain the underlying logic/assumptions/biases (from Step 2).
- Discuss the likely author's goal (from Step 3).
- Provide a clear verdict: "Unsupported Claim."
**Output Format:**
Present your findings as a structured report using Markdown. Use headings for each flagged claim and bullet points for the analysis (C, E, G, Verdict).
**Example Structure for a Flagged Claim:**
### Claim: "[The specific unsupported claim from the draft]"
* **Counterargument/Missing Evidence:** [Explanation of why it's unsupported, what's missing]
* **Explanation of Logic/Assumptions:** [Analysis of the claim's construction]
* **Author's Likely Goal:** [What the author might be trying to achieve]
* **Verdict:** Unsupported Claim
**Please provide the draft text you would like me to analyze.**
Get this written for your actual task
Paste what you are trying to do and the corpus will be matched against it directly. Free, no account, about ten seconds.
single retrieval pass
That number is low on purpose, and it is real. It is the raw similarity of one retrieval pass: no specialist read the paper, no judge compared anything, the first plausible match won.
one of which is this page
Picking the right one for a specific task is the work, and it is the work GetDecision does.
| This page | one technique, generic prompt |
| What you just ran | one technique matched to your wording, nothing verified |
| Full run | ten specialists read the papers in full, a judge ranks the top three for your task and shows its reasoning, generation on the model you pick, saved to your history |
See the top three for your taskTen specialists, a judge, and the reasoning shown. Free account, first run included.
Run the full analysis