🔬 Research
PRISMA 2020 Flow Narrative from Locked Screening Counts (No Invented DOIs) (Domain Pack)
Draft a PRISMA 2020 flow narrative from locked screening counts. No invented DOIs, study titles, or extra counts.
0Reviews
Prompt
Act as a systematic review methodologist who only uses locked screening counts. You draft a PRISMA 2020 flow narrative the counts already support. You do not invent DOIs, study titles, extra stage counts, or exclusion reasons not in Inputs. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Locked screening counts I lock: [Counts] - Review working title I lock: [Review] - Databases searched I lock (or NONE): [Databases] - Exclusion reason list I lock (or NONE): [Exclusions] - Included study list I lock (or NONE): [Included] - Section cap I lock: [Cap] - Words I must not use: [Banned] - What I must never invent (DOIs, titles, extra counts): [Never] - Output format: [Format] - Language: [Lang] Generate: 1. Honesty ledger: Counts figures, Review, Databases, Exclusions, Included, Cap, Lang. Forbidden: invented DOIs, study titles, extra counts. 2. PRISMA 2020 flow narrative: at most Cap stages using Counts only. Arithmetic must match Inputs. 3. Exclusion narrative: Exclusions only; else NOT IN INPUTS. 4. Included shelf: Included only. DOIs missing stay NOT IN INPUTS. 5. Refuse list: inventing DOIs, inventing PubMed IDs, inventing stage counts, inventing exclusion reasons. 6. Compliance pass: quote Banned and Never hits. Cut them. Print stage count. Format as Format. Constraints: - Flow narrative from Counts only. No invented DOIs. - Honor Cap. No emojis.
Instructions
Replace every [bracket] with your details before running. Works on ChatGPT, Claude, and Gemini.
Generated Output
This image was generated using the prompt above.
Examples
Example Input
Counts: identified 420 as pasted; duplicates removed 40 as pasted; title/abstract screened 380 as pasted; full text assessed 62 as pasted; included 18 as pasted; excluded at full text 44 as pasted. Review: Telemetry alerts in regional EMS as pasted Databases: MEDLINE as pasted; Scopus as pasted. Embase NOT IN INPUTS. Exclusions: wrong population 20 as pasted; not intervention 14 as pasted; incomplete outcomes 10 as pasted. Language exclude NONE. Included: study list NONE (titles/DOIs not provided) Cap: 6 Banned: invent DOI, invent PubMed ID, invent extra count Never: invent DOIs, invent study titles, invent extra counts Format: ledger + flow-narrative + exclusions + included-shelf + refuse + compliance Lang: English
Example Output
1. Ledger. Counts: identified 420; duplicates 40; screened 380; full text 62; included 18; full-text excluded 44. Review Telemetry alerts in regional EMS. Databases MEDLINE, Scopus; Embase NOT IN INPUTS. Exclusions wrong population 20, not intervention 14, incomplete outcomes 10; language NONE. Included list NONE. Cap 6. Lang English. Forbidden: invented DOIs, titles, extra counts. 2. Flow narrative (6/6 stages). Identification: 420 records as pasted across MEDLINE and Scopus as pasted. Embase: NOT IN INPUTS. Dedup: 40 duplicates removed as pasted -> 380 remain (420-40=380). Screening: 380 title/abstract as pasted. Eligibility: 62 full text assessed as pasted. Excluded full text: 44 as pasted. Included: 18 as pasted. Check: 62-44=18. 3. Exclusions. Wrong population 20 as pasted. Not intervention 14 as pasted. Incomplete outcomes 10 as pasted. Language exclude: NONE. Other reasons: NOT IN INPUTS. 4. Included shelf. Study titles: NOT IN INPUTS. DOIs: NOT IN INPUTS. PubMed IDs: NOT IN INPUTS. Do not fabricate citations. 5. Refuse. DOI invent: refused. PubMed invent: refused. Embase count invent: refused. extra stage invent: refused. study title invent: refused. 6. Compliance. Banned hits none. Format ledger+flow-narrative+exclusions+included-shelf+refuse+compliance. Gaps: Embase, language exclusions, included titles, DOIs, PubMed IDs. Missing-data policy: if a field was blank, write NOT IN INPUTS rather than guessing. Lock any tool version named in Inputs; if unnamed, write unknown. No invented testimonials, star ratings, or press logos. If legal, clinical, insurance, HR, education-plan, or veterinary content appears, add a one-line not-advice and de-identify banner. Quote banned-word hits and cut them. End with a gaps list of five bullets the user still owes you. Character and byte caps in the job are hard; print counts when relevant. Refuse to backfill DOIs, exam dumps, PHI, PII, or compensation promises not in Inputs.