Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

Inclusion Criteria Design SOP · yogsoth-ai | DecimalAI

yogsoth-ai/inclusion-criteria-design

SkillScore95Maintained

Define inclusion/exclusion criteria for systematic study selection in meta-analysis

researchby yogsoth-ai

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Research pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/yogsoth-ai-inclusion-criteria-design/SKILL.md

decimalai skills pull yogsoth-ai-inclusion-criteria-design --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/yogsoth-ai-inclusion-criteria-design/SKILL.md https://app.decimal.ai/s/yogsoth-ai-inclusion-criteria-design/SKILL.md

Raw file: /s/yogsoth-ai-inclusion-criteria-design/SKILL.md ↗ · machine-readable summary at /s/yogsoth-ai-inclusion-criteria-design.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 18d ago

Yes — +9% lift+16% tokens0% turns95% of whole cases pass with skill (21/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-10	✗→✓	▲ Improved	-14%	0%
case-12	✗→✓	▲ Improved	-16%	0%
case-22	✗→✓	▲ Improved	347%	0%
case-20	✓→✗	▼ Worse	267%	0%
case-01	✓→✓	＝ Same ✓	-35%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/10/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

yogsoth-ai / de-anthropocentric-research-engine

skills/inclusion-criteria-design/SKILL.md

View on GitHub ↗

License Apache-2.0Imported 8/10/2026Version v1

Skill contents

SKILL.md

Inclusion Criteria Design SOP

Define rigorous inclusion and exclusion criteria for study selection, ensuring reproducibility and minimizing selection bias.

When to Use

After PICO formulation is complete
Before beginning systematic search
When refining scope after pilot search reveals too many/few results

Input

pico_framework: The structured PICO/PECO output from pico-formulation
study_types: Acceptable study designs (RCT, cohort, case-control, cross-sectional, etc.)

Output

Complete eligibility criteria document with decision rules for borderline cases, suitable for independent screening by multiple reviewers.

Available SOPs

Optional, no fixed order; the final leaf is always a sop.

| SOP | When to use | | --- | --- | | spawn-agent | Spawn a customized CC subagent with full MCP tool access. Used by SOPs that declare execution: subagent. |

Illustrative Want to see what yogsoth-ai/inclusion-criteria-design would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Latest benchmark · publisher runVerified

9/2/2026, 16:37 UTC · 22 cases

Lift

+9%

95% vs 86%

Agent turns

0%

1.0 vs 1.0

Tokens

+16%

2,234.73 vs 1,922

fail → pass · 3pass → pass · 18pass → fail · 1

Per-case results (22)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-01 We are preparing a systematic review on SGLT2 inhibitors for heart failure and want to write the eligibility criteria document immediately. We have not yet formally defined the Population, Intervention, Comparator, and Outcome structure. Should we proceed directly to drafting the inclusion rules or complete the PICO framework first?	pass→pass	28,590	83,432	+192%	1	1	0%	1,967	1,281	-35%	0	0	—
case-02 Our team has completed the PICO framework for a review on mindfulness interventions for adolescent anxiety. A colleague suggests running the comprehensive database searches across PubMed and Embase before establishing the detailed inclusion/exclusion rules, arguing that seeing the search results first will help us write better rules. Is this the correct sequence?	pass→pass	42,252	49,709	+18%	1	1	0%	1,665	1,990	+20%	0	0	—
case-03 During a pilot search for a systematic review on hypertension management, our preliminary query returned over 45,000 preliminary results, which is far beyond our screening capacity. What action should be taken regarding the eligibility scope?	pass→pass	16,204	36,188	+123%	1	1	0%	2,160	2,483	+15%	0	0	—
case-04 A pilot search on a rare pediatric metabolic disorder yielded only 2 citations across three databases. How should the study eligibility scope be adjusted?	pass→pass	28,332	46,998	+66%	1	1	0%	2,306	2,961	+28%	0	0	—
case-05 We are initiating the inclusion criteria specification process for a clinical review. We have gathered clinical guidelines and background literature. What specific structured input framework from the prior stage is required to ground these criteria?	pass→pass	40,351	13,926	-65%	1	1	0%	1,444	716	-50%	0	0	—
case-06 When establishing inclusion criteria for a review on surgical techniques for ACL reconstruction, besides the population and intervention details, what specific study design classification parameters must be explicitly specified as inputs?	pass→pass	30,515	13,211	-57%	1	1	0%	2,353	2,081	-12%	0	0	—
case-07 What primary deliverable should be produced at the conclusion of the criteria design phase to allow dual independent reviewers to screen titles and abstracts without constant clarification meetings?	pass→pass	12,394	18,374	+48%	1	1	0%	1,586	1,200	-24%	0	0	—
case-08 In our systematic review protocol on cognitive behavioral therapy, we anticipate papers where patient age ranges overlap our target age cut-off of 18 (e.g., studies enrolling ages 15-22). What structural element must be added to the eligibility documentation to handle such ambiguous studies consistently?	pass→pass	15,510	7,946	-49%	1	1	0%	1,719	1,410	-18%	0	0	—
case-09 How should the final output of the eligibility criteria SOP be structured to ensure multiple independent screeners achieve high inter-rater agreement during title/abstract screening?	pass→pass	29,295	11,358	-61%	1	1	0%	2,676	2,442	-9%	0	0	—
case-10 We are defining inclusion criteria for a drug efficacy review where non-randomized studies present high risk of bias. How should study design constraints be declared in the eligibility input parameter?	fail→pass	16,955	15,429	-9%	1	1	0%	2,241	1,934	-14%	0	0	—
case-11 For a review evaluating rare adverse drug reactions occurring over 10 years, randomized trials are unfeasible. What study designs should be specified in the input parameters for eligibility?	pass→pass	21,326	11,657	-45%	1	1	0%	2,357	2,229	-5%	0	0	—
case-12 We just finalized our PICO elements for a review on remote patient monitoring. What is the immediate next step in the systematic review workflow before conducting the primary database search?	fail→pass	18,902	8,539	-55%	1	1	0%	1,604	1,349	-16%	0	0	—
case-13 Why is it critical to establish detailed, objective exclusion thresholds prior to screening rather than letting screeners judge study relevance intuitively during screening?	pass→pass	16,251	19,077	+17%	1	1	0%	2,103	2,438	+16%	0	0	—
case-14 A researcher suggests selecting studies as they read through search results based on overall scientific quality and interesting findings. What methodological risk does this introduce, and how does standard eligibility design address it?	pass→pass	11,079	19,777	+79%	1	1	0%	2,015	1,829	-9%	0	0	—
case-15 We are evaluating environmental exposure to microplastics rather than a therapeutic drug. Which variation of the structured framework input should be used to formulate eligibility criteria?	pass→pass	17,487	15,852	-9%	1	1	0%	1,602	1,736	+8%	0	0	—
case-16 When drafting the criteria document output, what feature ensures that two screeners working separately in different institutions reach identical inclusion decisions on marginal papers?	pass→pass	20,738	8,326	-60%	1	1	0%	1,884	1,430	-24%	0	0	—
case-17 Our protocol pilot search retrieved 12,000 papers for a team of two reviewers with a 2-week deadline. What step in the SOP does this signal?	pass→pass	16,200	6,894	-57%	1	1	0%	1,604	936	-42%	0	0	—
case-18 In drafting inclusion criteria for type 2 diabetes treatments, a team member wrote 'include adults with diabetes'. How should this criteria definition be improved for the SOP output?	pass→pass	16,598	11,380	-31%	1	1	0%	1,696	2,008	+18%	0	0	—
case-19 We are reviewing laparoscopic versus open appendectomy. Studies comparing laparoscopic appendectomy to robotic appendectomy were found in preliminary scans. How should this be addressed in the criteria document?	pass→pass	9,722	18,033	+85%	1	1	0%	1,696	1,885	+11%	0	0	—
case-20 We have a broad clinical question regarding whether exercise reduces depression in elderly patients. Convert this raw clinical question into a structured PICO format with explicit P, I, C, and O components.	pass→fail	10,252	23,171	+126%	1	1	0%	936	3,437	+267%	0	0	—
case-21 Translate our finalized inclusion criteria for pediatric asthma biological therapies into a full PubMed search string using MeSH terms and Boolean operators (AND/OR).	pass→pass	21,724	14,994	-31%	1	1	0%	2,784	2,955	+6%	0	0	—
case-22 Given data extracted from 5 included trials on statin therapy, calculate the pooled odds ratio and I-squared heterogeneity statistic using a random-effects model.	fail→pass	11,563	42,579	+268%	1	1	0%	1,886	8,434	+347%	0	0	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +9 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 86.4%
Pass rate with the skill: 95.5%
Cases attempted: 22
Cases counted toward lift: 22
Cases that regressed: 1
Skill version measured: v1
Run completed: 9/2/2026, 16:37 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-10

Prompt

We are defining inclusion criteria for a drug efficacy review where non-randomized studies present high risk of bias. How should study design constraints be declared in the eligibility input parameter?

Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.

Show both answers

SkillScore blends 4 signals1 of 4 complete

BenchmarkPass rate WITH skill on a verified run95

Live evalNeeds ≥10 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

Efficiency · separate axis (at equal correctness)+0% turns-4% tokens

Shown for transparency; not credited unless a verified run shows no correctness regression.

Backed by 1 of 4 signals — each added signal makes the score more robust.

Verified on gemini-3.6-flash · re-tested 17 days ago

Other measured skills in the registry, with their headline benchmark lift.