Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add hannesill/m4 --skill mimic-eicu-mappinggit clone --depth 1 https://github.com/hannesill/m4Wrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/hannesill/m4/mimic-eicu-mapping)<a href="https://agentmods.dev/skills/hannesill/m4/mimic-eicu-mapping"><img src="https://agentmods.dev/badge/skills/hannesill/m4/mimic-eicu-mapping/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/hannesill/m4/mimic-eicu-mapping"><img src="https://agentmods.dev/badge/skills/hannesill/m4/mimic-eicu-mapping.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00068 | $0.01411 |
| Opus 5.5 | $0.00027 | $0.00564 |
| Sonnet 5.5 | $0.00014 | $0.00282 |
| Haiku 4.5 | $0.00007 | $0.00141 |
Grade A, and why
mimic-eicu-mapping scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MIMIC-IV to eICU Mapping
eICU-CRD 2.0 is multi-center (208 US hospitals, 2014-2015); MIMIC-IV is single-center. Apply these rules to every eICU query.
Rules
- Check eICU is available. It is credentialed. On the default DuckDB backend, the
eicuentry ofm4 list-datasets --jsonneeds"prepared": true. On BigQuery (m4 doctor --jsonreports the backend and checks the credentials), it needsbigquery_available. Otherwise tell the user; never download it yourself. - Prefix every table with
eicu_crd.:eicu_crd.patient, notpatient. M4 exposes eICU only under this schema (physionet-data.eicu_crdon BigQuery). - Time is an integer offset. Each table has its own offset column, in minutes from this unit stay's admission (
labresultoffset,nursingchartoffset, ...); negative means before. There are no dates;hospitaldischargeyearis the only year. MIMICintime - 6 h .. intime + 24 hbecomesoffset BETWEEN -360 AND 1440. - Age is text,
'> 89'for older patients:CASE WHEN age = '> 89' THEN 91 ELSE TRY_CAST(age AS INTEGER) END(SAFE_CAST(age AS INT64)on BigQuery). 91 matches MIMIC-IV'sanchor_agefor patients over 89. - Clinical tables join on
patientunitstayid. Onepatientrow is one unit stay and covers MIMIC's patients, admissions and icustays; it also holds the coarser keys:patienthealthsystemstayid(hospital stay),uniquepid(patient), andhospitalid, which joinseicu_crd.hospital.
Identifiers
| MIMIC-IV | eICU |
|---|---|
subject_id |
uniquepid |
hadm_id |
patienthealthsystemstayid |
stay_id |
patientunitstayid |
| ICU stay order within admission | unitvisitnumber |
| (single hospital) | hospitalid → eicu_crd.hospital |
Tables
| MIMIC-IV | eICU | Offset; notes |
|---|---|---|
| patients, admissions, icustays | patient | hospitaladmitoffset, unitdischargeoffset, hospitaldischargeoffset; hospitaldischargestatus, unitdischargestatus |
| chartevents (monitor vitals) | vitalperiodic, vitalaperiodic | observationoffset; heartrate, respiration, sao2, systemicmean; noninvasive BP in vitalaperiodic |
| chartevents (nurse-charted, GCS) | nursecharting | nursingchartoffset; text nursingchartvalue, keyed by nursingchartcelltypevallabel + nursingchartcelltypevalname |
| labevents | lab | labresultoffset; free-text labname (list with SELECT labname, COUNT(*) FROM eicu_crd.lab GROUP BY 1), numeric labresult |
| outputevents | intakeoutput | intakeoutputoffset; urine by cellpath |
| inputevents | infusiondrug | infusionoffset |
| prescriptions, emar | medication | drugstartoffset, drugstopoffset |
| microbiologyevents | microlab | culturetakenoffset |
| diagnoses_icd | diagnosis | diagnosisoffset; text diagnosisstring; icd9code coverage varies |
| (none) | admission diagnosis | patient.apacheadmissiondx, apachepredvar.admitdiagnosis, admissiondx |
| procedureevents (ventilation) | respiratorycare, treatment | respcarestatusoffset, treatmentoffset |
| (none) | apachepatientresult, apacheapsvar, apachepredvar | APACHE IV/IVa scores and predictions (one row per apacheversion); physiology inputs; predictors |
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 7d ago Changed · -117 lines · +29 tokens per session 898648794915
- 1mo ago Changed · +1 lines f03c3d8fcf16
- 1mo ago First seen · 200 lines · 39 tokens per session scan A 9db5c6fde7d2
mimic-eicu-mapping is a skill published in the GitHub repository hannesill/m4 (46 stars, last pushed 2d ago), licensed MIT. It adds 68 tokens to every session and 1,411 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
migration-review
Review database migration files when a change adds or modifies paths under migrations/. Use it before merge to collect forward, rollback, locking, and data-safety evidence.
azure-resource-manager-mysql-dotnet
Azure MySQL Flexible Server SDK for .NET. Database management for MySQL Flexible Server deployments. Use for creating servers, databases, firewall rules, configurations, backups, and high availability. Triggers: "MySQL", "MySqlFlexibleServer", "MySQL Flexible Server", "Azure Database for MySQL", "MySQL database…
azure-resource-manager-postgresql-dotnet
Azure PostgreSQL Flexible Server SDK for .NET. Database management for PostgreSQL Flexible Server deployments. Use for creating servers, databases, firewall rules, configurations, backups, and high availability. Triggers: "PostgreSQL", "PostgreSqlFlexibleServer", "PostgreSQL Flexible Server", "Azure Database for…
azure-resource-manager-sql-dotnet
Azure Resource Manager SDK for Azure SQL in .NET. Use for MANAGEMENT PLANE operations: creating/managing SQL servers, databases, elastic pools, firewall rules, and failover groups via Azure Resource Manager. NOT for data plane operations (executing queries) - use Microsoft.Data.SqlClient for that. Triggers: "SQL…
azure-data-tables-java
Build table storage applications with Azure Tables SDK for Java. Use when working with Azure Table Storage or Cosmos DB Table API for NoSQL key-value data, schemaless storage, or structured data at scale.
azure-cosmos-java
Azure Cosmos DB SDK for Java. NoSQL database operations with global distribution, multi-model support, and reactive patterns. Triggers: "CosmosClient java", "CosmosAsyncClient", "cosmos database java", "cosmosdb java", "document database java".