Resource
Agent Task Template
A reusable AI agent task template with variables for objective, context, available tools, constraints, success criteria, failure handling, and output format.
Research Comparison Template
A reusable multi-source comparison template with variables for sources, evaluation criteria, audience, decision context, priority criteria, and output format.
Overview
Comparing multiple sources without structure produces averaging — strong sources dominate, nuanced disagreements disappear, and the output is a consensus that none of the sources actually support. This template forces per-source extraction before synthesis, names criteria explicitly, and requires the output to surface contradictions rather than resolve them. The result is a comparison your decision-makers can trust rather than one that just feels comprehensive.
How to use this resource
-
Define evaluation criteria before reading the sources
The evaluationCriteria variable should be set before you read the sources. Criteria defined after reading tend to be shaped by what you found, not by what the decision needs.
-
Fill in the template variables
Open in Prompt Template Builder. Variables: sources, evaluationCriteria, audience, decisionContext, priorityCriteria, outputFormat, caveats.
-
Label sources consistently
Give each source a clear identifier in the sources variable — title, author, or URL. The model needs these to attribute claims. Unlabeled sources produce unattributed conclusions.
-
Focus review on contradictions and gaps
The most valuable output section is the contradictions and evidence gaps. That is where you will find what the available research does not support, which is often more decision-relevant than what it does.
Why This Works
- Per-source extraction before synthesis prevents one strong source from anchoring the entire comparison — weaker sources get the same structured treatment
- Named attribution for every claim makes it easy to trace a conclusion back to its evidence when the decision is challenged
- Explicit contradiction flagging surfaces disagreements that averaging would erase — genuine expert disagreement is information, not noise
- Evidence gap tracking makes the boundaries of what you actually know visible before you make the decision
Best for
- Decisions where source attribution matters — when you need to know which source supports which conclusion
- Topics where expert disagreement is meaningful rather than a rounding error
- Comparisons where criteria need to be consistent across all sources
- Pre-decision analysis where you need a defensible synthesis rather than a personal judgment
Not for
- Single-source analysis — use a summarization template instead
- Real-time research where you cannot paste sources in full
- Sources requiring specialist domain knowledge to evaluate fairly without bias
Use cases
- Comparing vendor proposals or tool evaluations with consistent criteria
- Synthesizing competing research papers before writing a literature review section
- Consolidating stakeholder or user research findings across multiple interview sessions
- Reviewing multiple analyst reports on the same market or technology
FAQ
How does the template stop strong sources from dominating the comparison?
By forcing per-source extraction before any synthesis: each source gets the same structured treatment — key claims with evidence, methodology, explicit caveats — so a weaker source isn't drowned out by a stronger one anchoring the whole read. Only after that does cross-source synthesis run. It's a fill-in-the-variables template you complete before running it in your own assistant.
Why does it surface contradictions instead of resolving them into one answer?
Because genuine expert disagreement is information, not noise. The synthesis section requires naming contradictions with the specific sources involved and listing evidence gaps none of the sources address, rather than averaging them into a consensus none actually support. Speculative conclusions get a [speculative] flag, and every claim must be attributed to its source, not blended anonymously.
Which variables do I need to fill before running the comparison?
Open it in Prompt Template Builder and fill {{sources}}, {{evaluationCriteria}}, {{audience}}, {{decisionContext}}, {{priorityCriteria}}, {{outputFormat}}, and {{caveats}}. The workflow stresses setting evaluationCriteria before reading the sources and giving each source a clear identifier in {{sources}} — the model needs those labels to attribute claims, and unlabeled sources produce unattributed conclusions.
More resources from Prompt Template Builder
Resource
Blog Post Template
A reusable long-form content template with variables for topic, audience, tone, target keyword, content angle, outline, and call to action.
Resource
Pull Request Review Template
A reusable PR review prompt template with variables for repository, branch, change summary, changed files, review criteria, and risk areas.
Resources that pair well
Resource
Bug Triage Assistant
Convert scattered bug notes, Slack messages, or user complaints into structured engineering tasks with reproduction steps, severity, and root cause hypothesis.
Resource
API Request Variable Builder
Turns a reusable API request prompt into a clean variable set covering endpoint, method, auth, payload, response format, and error handling.
Related tools
Tool
Prompt Template Builder
Build reusable prompt templates with {{variable}} placeholders, live preview, and export.