Assumption Auditing Chain

validated scientific chain v1.0.0 cc-by-sa

Block ID: 9c95c15f-30fb-478c-a2c4-1d0afa05db01

Community-contributed block. PromptDNA makes no guarantee of output quality or fitness for purpose. User assumes all responsibility for use.

Template

Audit the assumptions underlying the analysis, model, or argument in {analysis_context}, since conclusions are only as sound as their premises. (1) Surface the assumptions, including the tacit ones: read the analysis and list not only what it states but what it takes for granted — linearity, independence, stationarity, representativeness, ceteris paribus, a stable relationship holding out of sample. The unstated assumptions are the dangerous ones. (2) Classify each: is it a simplifying assumption (known to be false but useful), an empirical assumption (a claim about the world that could be checked), or a structural assumption baked so deep that violating it invalidates the whole approach? (3) For each assumption, ask what happens if it's wrong — does the conclusion degrade gracefully or collapse? Rank assumptions by how much the conclusion depends on them, not by how questionable they seem in isolation. (4) Check which assumptions are actually testable against available data, and test the load-bearing ones. (5) Identify assumptions imported from a context where they held into one where they may not — the most common silent failure. (6) State which assumptions the conclusion can survive being wrong about and which it cannot, so the reader knows exactly where to press.

Audit the assumptions underlying the analysis, model, or argument in {analysis_context}, since conclusions are only as sound as their premises. (1) Surface the assumptions, including the tacit ones: read the analysis and list not only what it states but what it takes for granted — linearity, independence, stationarity, representativeness, ceteris paribus, a stable relationship holding out of sample. The unstated assumptions are the dangerous ones. (2) Classify each: is it a simplifying assumption (known to be false but useful), an empirical assumption (a claim about the world that could be checked), or a structural assumption baked so deep that violating it invalidates the whole approach? (3) For each assumption, ask what happens if it's wrong — does the conclusion degrade gracefully or collapse? Rank assumptions by how much the conclusion depends on them, not by how questionable they seem in isolation. (4) Check which assumptions are actually testable against available data, and test the load-bearing ones. (5) Identify assumptions imported from a context where they held into one where they may not — the most common silent failure. (6) State which assumptions the conclusion can survive being wrong about and which it cannot, so the reader knows exactly where to press.

Variables

Name	Type	Required	Trust level
analysis_context		yes

assumptionsmodel-critiqueepistemicschain-of-thought

Ratings

0.0

Overall (0)

0.0

Accuracy

0.0

Consistency

0.0

Clarity

0.0

Efficiency

Benchmarks

Not yet self-validated against any benchmark. Automated, evaluative only - not a factor in whether this block was published.

Submitted by James P FounderMod via mcp · 2026-07-18