R0049/2026-03-31/Q001-H2¶
Statement¶
No published AI/LLM system prompt exists that implements any form of analytical rigor framework for research.
Status¶
Eliminated. Multiple partial implementations were found, including published source code for LLM-based structured analytic techniques and academic papers describing prompt engineering for systematic review tasks.
Supporting Evidence¶
| Evidence | Summary |
|---|---|
| SRC03-E01 | Major prompt libraries do not contain comprehensive research-methodology prompts (but this only shows gaps in collections, not absence everywhere) |
Contradicting Evidence¶
| Evidence | Summary |
|---|---|
| SRC04-E01 | Roberts published LLM implementations of ACH, Starbursting, and Key Assumptions Check with source code on GitHub |
| SRC02-E01 | Framework Chain-of-Thought is a published prompting approach implementing systematic review framework reasoning |
| SRC01-E01 | PRISMA-trAIce provides a checklist for AI-assisted systematic reviews, indicating active framework development |
Reasoning¶
The evidence clearly shows that partial implementations exist. While none achieve the comprehensive coverage described in H1, the existence of published prompts implementing individual structured analytic techniques and systematic review screening methodologies eliminates the hypothesis that no such work exists.
Relationship to Other Hypotheses¶
Directly contradicted by evidence supporting H1 and H3. The distinction between H1 and H3 is more informative than H2.