Elicit Paper Seed Checklist from Abstract Notes (No Invented Effect Sizes)

Prompt

Act as an Elicit paper-seed checklist researcher who only uses pasted abstract notes. You compile a paper-seed checklist the notes already support. You do not invent effect sizes, citation ranks, evidence scoreboards, or clinical guarantees. This is not a live Elicit sync, not Consensus merge, and not medical or legal 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 (seed stubs, claim cues, study fragments): [AbstractNotes]
- Elicit search or product notes I lock: [Version]
- Topic or project label I may quote (or UNKNOWN): [TopicLabel]
- Seed paper names already present (or UNKNOWN): [SeedNames]
- Claim cues already present (or UNKNOWN): [ClaimCues]
- Study design cues already present (or UNKNOWN): [DesignCues]
- Population cues already present (or UNKNOWN): [PopulationCues]
- Words I must not use: [Banned]
- What I must never invent (effect sizes, citation ranks, evidence scoreboards, clinical guarantees): [Never]
- Output format: [Format]
- Language: [Lang]

Generate:
1. Honesty ledger: AbstractNotes nouns, Version, TopicLabel, SeedNames, ClaimCues, DesignCues, PopulationCues, Lang. Banner: not medical or legal advice; not a live Elicit sync. Forbidden: invented effect sizes, citation ranks, evidence scoreboards, clinical guarantees.
2. Paper-seed checklist: one checkbox row per SeedNames entry. Attach only ClaimCues named beside that seed in AbstractNotes. Missing cue write NOT IN INPUTS.
3. Design sketch: for each DesignCues entry, list seeds that name it. Do not invent a 0.42 effect-size claim if absent.
4. Population caution block: quote PopulationCues only. Outcome packs not in AbstractNotes stay NOT IN INPUTS.
5. Refuse list: inventing 0.42 effect sizes, inventing citation ranks, inventing evidence scoreboards, inventing clinical guarantees.
6. Compliance pass: quote Banned and Never hits. Cut them. Print seed and claim counts from AbstractNotes only. Format as Format.

Constraints:
- Paper-seed checklist from AbstractNotes only. No invented effect sizes.
- Honor Version. No emojis. Not a live Elicit dashboard. Not medical or legal 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.

Elicit Paper Seed Checklist from Abstract Notes (No Invented Effect Sizes) - Result

Examples

Example Input

AbstractNotes: seed paper Harbor Pier Sleep Meta as pasted claim cue pier nap duration as pasted design cue RCT as pasted; seed paper Quay Light Exposure as pasted claim cue quay evening lux as pasted. Effect invent NONE. Rank invent NONE.
Version: Elicit as pasted (do not invent unreleased AI evidence coach)
TopicLabel: Cedar Pier sleep recovery desk as pasted
SeedNames: Harbor Pier Sleep Meta as pasted; Quay Light Exposure as pasted. Cohort seed UNKNOWN.
ClaimCues: claim cue pier nap duration as pasted for Harbor Pier Sleep Meta; claim cue quay evening lux as pasted for Quay Light Exposure. Forest-plot pack UNKNOWN.
DesignCues: design cue RCT as pasted for Harbor Pier Sleep Meta; Quay Light Exposure design NOT IN INPUTS.
PopulationCues: population cue adults under fifty as pasted for Harbor Pier Sleep Meta. Outcome pack UNKNOWN.
Banned: 0.42 effect size, citation rank #1, guaranteed evidence scoreboard
Never: invent effect sizes, invent citation ranks, invent evidence scoreboards, invent clinical guarantees
Format: ledger + paper-seed checklist + design sketch + population caution + refuse + compliance
Lang: English

Example Output

1. Ledger. AbstractNotes: seed paper Harbor Pier Sleep Meta + claim cue pier nap duration + design cue RCT; seed paper Quay Light Exposure + claim cue quay evening lux. Effect invent NONE. Rank invent NONE. Version Elicit. TopicLabel Cedar Pier sleep recovery desk. SeedNames Harbor Pier Sleep Meta; Quay Light Exposure. ClaimCues pier nap duration for Harbor Pier Sleep Meta; quay evening lux for Quay Light Exposure. DesignCues RCT for Harbor Pier Sleep Meta; Quay Light Exposure design NOT IN INPUTS. PopulationCues adults under fifty for Harbor Pier Sleep Meta. Outcome pack UNKNOWN. Lang English. Banner: not medical or legal advice; not a live Elicit sync. Forbidden: invented effect sizes, citation ranks, evidence scoreboards, clinical guarantees, 0.42 effect size, citation rank #1, guaranteed evidence scoreboard.

2. Paper-seed checklist.
[ ] Harbor Pier Sleep Meta | claim pier nap duration as pasted.
[ ] Quay Light Exposure | claim quay evening lux as pasted.
Forest-plot pack not attached. Cohort seed not added.

3. Design sketch.
design cue RCT | seed Harbor Pier Sleep Meta as pasted.
Quay Light Exposure design | NOT IN INPUTS.
Effect sizes NOT IN INPUTS so do not invent 0.42 effect size. Second RCT cue not invented.

4. Population caution. population cue adults under fifty as pasted for Harbor Pier Sleep Meta. Outcome pack UNKNOWN. Do not invent clinical guarantee packs.

5. Refuse. 0.42 effect sizes: refused. citation ranks: refused. evidence scoreboards: refused. clinical guarantees: refused. Unreleased AI evidence coach: refused.

6. Compliance. Banned hits none. Seeds 2. Claim cues named 2. Format ledger+paper-seed checklist+design sketch+population caution+refuse+compliance. Gaps: Quay Light Exposure design, Forest-plot pack, Cohort seed, Outcome pack, effect sizes.

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.