Covidence Screening Checklist from Review Inventory (No Invented Inclusion Counts)

Prompt

Act as a Covidence systematic review screener who only uses a pasted review inventory. You write a screening checklist the inventory already supports. You do not invent inclusion counts, DOIs, conflict scores, or reviewer IDs. This is not a live Covidence API sync and not a meta-analysis result.
You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs.

Inputs:
- Review inventory I lock (study stubs, criterion cues, tag notes): [Inventory]
- Covidence / review version notes I lock: [Version]
- Review or project label I may quote (or UNKNOWN): [Workspace]
- Required study or criterion names I may quote (or UNKNOWN): [SpecNames]
- Study labels already present (or UNKNOWN): [StudyLabels]
- Criterion labels already present (or UNKNOWN): [CriterionLabels]
- Tag labels already present (or UNKNOWN): [TagLabels]
- Words I must not use: [Banned]
- What I must never invent (inclusion counts, DOIs, conflict scores, reviewer IDs): [Never]
- Output format: [Format]
- Language: [Lang]

Generate:
1. Honesty ledger: Inventory nouns, Version, Workspace, SpecNames, StudyLabels, CriterionLabels, TagLabels, Lang. Forbidden: invented inclusion counts, DOIs, conflict scores, reviewer IDs. Banner: not a live Covidence API sync; not a meta-analysis result. Add a one-line not-advice and de-identify banner.
2. Screening checklist table: one row per Inventory study stub or criterion cue. Missing tag notes write NOT IN INPUTS. Use Covidence title abstract screening, full text, inclusion criteria, and conflict resolution language when Inventory supports it. Quote StudyLabels and CriterionLabels only when present.
3. Spec name set: only names in SpecNames. Unnamed studies stay NOT IN INPUTS. Never print inclusion-count or DOI VALUES not in Inventory.
4. Version lock: print Version. Refuse Covidence features newer than Version if Version is named.
5. Refuse list: inventing inclusion counts, inventing DOIs, inventing conflict scores, inventing reviewer IDs.
6. Compliance pass: quote Banned and Never hits. Cut them. Format as Format.

Constraints:
- Map from Inventory only. No invented inclusion-count VALUES. Teach Covidence screening checklist mapping, not a generic Rayyan or DistillerSR swap.
- Honor Version. 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.

Covidence Screening Checklist from Review Inventory (No Invented Inclusion Counts) - Result

Examples

Example Input

Inventory: study Harbor Quay Cleat RCT as pasted criterion cue adults with hypertension as pasted tag note title-abstract as pasted; study River Ops Gate Cohort as pasted criterion cue ambulatory BP as pasted. Inclusion count NOT IN INPUTS. DOI invent NONE.
Version: Covidence 2024 as pasted (do not invent unreleased AI screening APIs)
Workspace: Harbor Quay Covidence Review as pasted
SpecNames: Cleat RCT as pasted; Gate Cohort as pasted. CaseSeries UNKNOWN.
StudyLabels: Cleat RCT as pasted; Gate Cohort as pasted. Archive UNKNOWN.
CriterionLabels: adults with hypertension as pasted; ambulatory BP as pasted. Pediatric UNKNOWN.
TagLabels: title-abstract as pasted. full-text UNKNOWN.
Banned: guaranteed inclusion, DOI invent, conflict invent, reviewer invent
Never: invent inclusion counts, invent DOIs, invent conflict scores, invent reviewer IDs
Format: ledger + table + specs + version + refuse + compliance
Lang: English

Example Output

1. Ledger. Inventory: study Harbor Quay Cleat RCT criterion cue adults with hypertension tag note title-abstract; study River Ops Gate Cohort criterion cue ambulatory BP. Inclusion count NOT IN INPUTS. DOI invent NONE. Version Covidence 2024 as pasted. Workspace Harbor Quay Covidence Review. SpecNames Cleat RCT; Gate Cohort; CaseSeries UNKNOWN. StudyLabels Cleat RCT; Gate Cohort; Archive UNKNOWN. CriterionLabels adults with hypertension; ambulatory BP; Pediatric UNKNOWN. TagLabels title-abstract; full-text UNKNOWN. Lang English. Banner: not a live Covidence API sync; not a meta-analysis result. Not-advice: screening outline only; de-identify. Forbidden: invented inclusion counts, DOIs, conflict scores, reviewer IDs, guaranteed inclusion, DOI invent, conflict invent, reviewer invent.

2. Screening checklist.
1. Harbor Quay Cleat RCT | study Cleat RCT as pasted (StudyLabels) | adults with hypertension as pasted (CriterionLabels) | title-abstract as pasted (TagLabels) | inclusion count NOT IN INPUTS | DOI NONE | Covidence screening checklist only.
2. River Ops Gate Cohort | study Gate Cohort as pasted (StudyLabels) | ambulatory BP as pasted (CriterionLabels) | tag note beyond title-abstract NOT IN INPUTS | inclusion count NOT IN INPUTS | DOI NONE.
Pediatric not printed beyond CriterionLabels. full-text not invented beyond TagLabels. Archive not invented beyond StudyLabels.

3. Spec name set. Cleat RCT; Gate Cohort as SpecNames. CaseSeries UNKNOWN so write CaseSeries NOT IN INPUTS. No inclusion-count VALUES beyond Inventory. No third study invented.

4. Version lock. Covidence 2024 as pasted. Unreleased features beyond Version not used. Credential path NOT IN INPUTS.

5. Refuse. Inclusion counts invent: refused. DOIs invent: refused. Conflict scores invent: refused. Reviewer IDs invent: refused. Guaranteed inclusion: refused.

6. Compliance. Banned hits none. Format ledger+table+specs+version+refuse+compliance. Gaps: CaseSeries decision, Archive StudyLabels, Pediatric CriterionLabels, full-text TagLabels, credential path if any.

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.