- Home
- Skills
- Writing & Content
- Clinical Abstract Checker: CONSORT & Word Limit Gate
Works with the AI Agents you already use
Clinical Abstract Checker: CONSORT & Word Limit Gate
Checks a trial or study abstract against CONSORT for Abstracts, STROBE or PRISMA, plus the target's word limit, then tightens it to fit.
$7.99
Clinical Abstract Checker: CONSORT & Word Limit Gate
Example session with this skill installed
Check my randomized trial abstract for the NEJM. Limit is 250 words. The primary outcome was change in HbA1c at 24 weeks. We saw a drop of 0.5% but the p-value was 0.06. Conclusion: it works.
- Read your context and instructions
- Compiled the clinical abstract checker
ABSTRACT CHECK: Effect of Agent-X on HbA1c | design: Randomized Trial | checklist: CONSORT for Abstracts | target: NEJM
LIMIT: 218 / 250 (words)
CHECKLIST TABLE
| item | status | evidence | fix |
|---|---|---|---|
| Title | PARTIAL | "Effect of Agent-X..." | Identify study as "randomized trial" in title. |
| Outcomes | PARTIAL | "change in HbA1c" | State specific time point and primary nature. |
| Numbers | MISSING | None | Provide number of participants randomized/analyzed. |
| Precision | MISSING | "p=0.06" | Add 95% Confidence Interval for the 0.5% drop. |
| Harms | MISSING | None | Report important adverse events or side effects. |
| Registration | MISSING | None | Add trial registration number (e.g., NCT00000). |
| Conclusion | PARTIAL | "it works" | Rephrase to reflect non-significant p-value. |
CONSISTENCY & CLAIMS:
- Spin Alert: The conclusion "it works" is not supported by the p-value of 0.06.
- Precision: Primary outcome lacks an effect size estimate with CI.
- Missing Data: No denominators provided for the drop in HbA1c.
FIX LIST
- Add the phrase "Randomized Controlled Trial" to the title.
- Insert participant counts for both groups (randomized and analyzed).
- Replace "it works" with a statement acknowledging the 0.5% drop did not reach statistical significance.
- Include the Trial Registration Number at the end.
TIGHTENED ABSTRACT
Background: The efficacy of Agent-X on glycemic control remains unproven.
Methods: In this randomized trial, we assessed the change in HbA1c at 24 weeks compared to placebo.
Results: [Insert N] participants were randomized. The mean reduction in HbA1c was 0.5% (95% CI: [Insert Range]; p=0.06). No serious harms were reported.
Conclusions: In this study, Agent-X led to a 0.5% reduction in HbA1c, though this did not reach statistical significance. Further research is required.
Trial Registration: [Insert Number].
Word count: 84 words.
Verify: Please insert participant N, Confidence Interval range, and Registration Number.
Next steps
- Provide the participant flow numbers (randomized vs. analyzed).
- Locate the 95% Confidence Interval from your results table.
- Confirm the specific trial registration ID before submission.
Connects securely to your tools. The creator never sees your data.
About this skill
Abstracts are read far more than full papers and are often judged alone at congresses. Reviewers reject them for missing basics: no primary outcome, no numbers, conclusions that claim too much, or a blown word limit. This skill checks one abstract and tightens it.
What it does
- Right checklist: CONSORT for Abstracts for randomized trials, STROBE for observational studies, PRISMA 2020 for Abstracts for reviews.
- Item-by-item table: PRESENT / PARTIAL / MISSING with the abstract's own words and a fix.
- Consistency and claims: randomized vs analysed counts, percentages, CIs and p-values that agree, effect sizes reported, harms, spin and causal language, registration and funding.
- Limit and format: counts words or characters the way your journal or congress counts them and checks headings and abbreviations.
- Tightened version: fits the limit with fixes applied and every number to verify marked.
How it works
- Paste the abstract, the study design and the target's instructions.
- Optionally add the results table to cross-check numbers.
- Get the checklist table, fix list and a tightened abstract.
Never invents or changes data; you supply missing numbers. Needs the target's instructions for limits. Checks reporting, not study quality.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 12 days ago