Consensus Evidence Quote Checklist from Abstract Notes (No Invented P Values)

Prompt

Act as a Consensus evidence-quote checklist analyst who only uses pasted abstract notes. You compile an evidence-quote checklist the notes already support. You do not invent p values, effect sizes, citation scoreboards, or study ranks. This is not a live Consensus sync, not Semantic Scholar merge, and not clinical advice.
You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs.

Inputs:
- Abstract notes I lock (claim stubs, quote cues, study fragments): [AbstractNotes]
- Consensus project or version notes I lock: [Version]
- Desk or topic label I may quote (or UNKNOWN): [TopicLabel]
- Claim labels already present (or UNKNOWN): [ClaimLabels]
- Quote cues already present (or UNKNOWN): [QuoteCues]
- Study cues already present (or UNKNOWN): [StudyCues]
- Limit cues already present (or UNKNOWN): [LimitCues]
- Words I must not use: [Banned]
- What I must never invent (p values, effect sizes, citation scoreboards, study ranks): [Never]
- Output format: [Format]
- Language: [Lang]

Generate:
1. Honesty ledger: AbstractNotes nouns, Version, TopicLabel, ClaimLabels, QuoteCues, StudyCues, LimitCues, Lang. Banner: not clinical advice; not a live Consensus sync. Forbidden: invented p values, effect sizes, citation scoreboards, study ranks.
2. Evidence-quote checklist: one checkbox row per ClaimLabels entry. Attach only QuoteCues named beside that claim in AbstractNotes. Missing cue write NOT IN INPUTS.
3. Study sketch: for each StudyCues entry, list claims that name it. Do not invent a p=0.01 claim if absent.
4. Limit caution block: quote LimitCues only. Meta packs not in AbstractNotes stay NOT IN INPUTS.
5. Refuse list: inventing p=0.01 values, inventing effect sizes, inventing citation scoreboards, inventing study ranks.
6. Compliance pass: quote Banned and Never hits. Cut them. Print claim and quote counts from AbstractNotes only. Format as Format.

Constraints:
- Evidence-quote checklist from AbstractNotes only. No invented p values or effect sizes.
- Honor Version. No emojis. Not a live Consensus dashboard. Not clinical advice.

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.

Consensus Evidence Quote Checklist from Abstract Notes (No Invented P Values) - Result

Examples

Example Input

AbstractNotes: claim label Harbor Quay Sleep Latency as pasted quote cue mean latency drop noted as pasted study cue small RCT snippet as pasted; claim label Pier Deck Caffeine Timing as pasted quote cue evening cutoff noted as pasted. P invent NONE. Effect invent NONE.
Version: Consensus as pasted (do not invent unreleased AI meta judge)
TopicLabel: Cedar Pier sleep desk as pasted
ClaimLabels: Harbor Quay Sleep Latency as pasted; Pier Deck Caffeine Timing as pasted. Melatonin claim UNKNOWN.
QuoteCues: quote cue mean latency drop noted as pasted for Harbor Quay Sleep Latency; quote cue evening cutoff noted as pasted for Pier Deck Caffeine Timing. Dose quote UNKNOWN.
StudyCues: study cue small RCT snippet as pasted for Harbor Quay Sleep Latency. Pier Deck Caffeine Timing study NOT IN INPUTS.
LimitCues: limit cue short follow-up as pasted for Harbor Quay Sleep Latency. Pier Deck Caffeine Timing limit NOT IN INPUTS.
Banned: p=0.01 invented, large effect size guaranteed, citation scoreboard #1
Never: invent p values, invent effect sizes, invent citation scoreboards, invent study ranks
Format: ledger + evidence-quote checklist + study sketch + limit caution + refuse + compliance
Lang: English

Example Output

1. Ledger. AbstractNotes: claim label Harbor Quay Sleep Latency + quote cue mean latency drop noted + study cue small RCT snippet; claim label Pier Deck Caffeine Timing + quote cue evening cutoff noted. P invent NONE. Effect invent NONE. Version Consensus. TopicLabel Cedar Pier sleep desk. ClaimLabels Harbor Quay Sleep Latency; Pier Deck Caffeine Timing. QuoteCues mean latency drop noted for Harbor Quay Sleep Latency; evening cutoff noted for Pier Deck Caffeine Timing. StudyCues small RCT snippet for Harbor Quay Sleep Latency. Pier Deck Caffeine Timing study NOT IN INPUTS. LimitCues short follow-up for Harbor Quay Sleep Latency; Pier Deck Caffeine Timing limit NOT IN INPUTS. Lang English. Banner: not clinical advice; not a live Consensus sync. Forbidden: invented p values, effect sizes, citation scoreboards, study ranks, p=0.01 invented, large effect size guaranteed, citation scoreboard #1.

2. Evidence-quote checklist.
[ ] Harbor Quay Sleep Latency | quote mean latency drop noted as pasted | study small RCT snippet as pasted.
[ ] Pier Deck Caffeine Timing | quote evening cutoff noted as pasted | study NOT IN INPUTS.
Dose quote not attached. Melatonin claim not added.

3. Study sketch.
study cue small RCT snippet | claim Harbor Quay Sleep Latency as pasted.
Pier Deck Caffeine Timing study NOT IN INPUTS so do not invent a second study. P values NOT IN INPUTS so do not invent p=0.01.

4. Limit caution. limit cue short follow-up as pasted for Harbor Quay Sleep Latency. Pier Deck Caffeine Timing limit NOT IN INPUTS. Do not invent meta packs.

5. Refuse. p=0.01 values: refused. effect sizes: refused. citation scoreboards: refused. study ranks: refused. Unreleased AI meta judge: refused.

6. Compliance. Banned hits none. Claims 2. Quote cues named 2. Format ledger+evidence-quote checklist+study sketch+limit caution+refuse+compliance. Gaps: Pier Deck Caffeine Timing study, Pier Deck Caffeine Timing limit, Dose quote, Melatonin claim, P values.

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.