Ready to supercharge your AI agents?

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

Chip Clonal Hematopoiesis Agent · FreedomIntelligence | DecimalAI

Chip Clonal Hematopoiesis Agent

SkillScore96Maintained

chip-clonal-hematopoiesis-agent

AI-powered clonal hematopoiesis of indeterminate potential (CHIP) detection, risk stratification, and cardiovascular/malignancy risk prediction using genomic and clinical data.

scienceby FreedomIntelligence

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

Automatic

Clinical / Healthcare pack

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

pip install decimalai

2 · Get the skill — writes .claude/skills/chip-clonal-hematopoiesis-agent/SKILL.md

decimalai skills pull chip-clonal-hematopoiesis-agent --out .claude/skills/

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

curl -fsSL --create-dirs -o .claude/skills/chip-clonal-hematopoiesis-agent/SKILL.md https://app.decimal.ai/s/chip-clonal-hematopoiesis-agent/SKILL.md

Raw file: /s/chip-clonal-hematopoiesis-agent/SKILL.md ↗ · machine-readable summary at /s/chip-clonal-hematopoiesis-agent.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

24 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 50d ago

Yes — +21% lift · at higher cost+54% tokens0% turns96% of whole cases pass with skill (23/24)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-02	✗→✓	▲ Improved	6%	0%
case-03	✗→✓	▲ Improved	-17%	0%
case-06	✗→✓	▲ Improved	78%	0%
case-07	✗→✓	▲ Improved	46%	0%
case-09	✗→✓	▲ Improved	75%	0%

Showing 5 of 24 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/1/2026

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

FreedomIntelligence / OpenClaw-Medical-Skills

skills/chip-clonal-hematopoiesis-agent/SKILL.md

View on GitHub ↗

Imported 8/1/2026Version v2

Skill contents

SKILL.md

<!--

COPYRIGHT NOTICE

This file is part of the "Universal Biomedical Skills" project.

Copyright (c) 2026 MD BABU MIA, PhD <md.babu.mia@mssm.edu>

All Rights Reserved.

This code is proprietary and confidential.

Unauthorized copying of this file, via any medium is strictly prohibited.

Provenance: Authenticated by MD BABU MIA

-->

CHIP Clonal Hematopoiesis Agent

The CHIP Clonal Hematopoiesis Agent provides comprehensive detection and risk stratification of clonal hematopoiesis of indeterminate potential (CHIP). It identifies clonal mutations in blood cells, assesses risk of progression to myeloid malignancy, and predicts cardiovascular disease risk, integrating with the CHIC machine learning framework for CBC-based screening.

When to Use This Skill

When detecting CHIP mutations from blood sequencing data.
For stratifying risk of progression to MDS/AML.
To assess CHIP-associated cardiovascular disease risk.
When filtering CHIP variants from tumor liquid biopsy.
For population-level CHIP screening and research.

Core Capabilities

CHIP Detection: Identify clonal mutations with VAF >2%.

Risk Stratification: Predict myeloid malignancy progression risk.

CVD Risk Assessment: Estimate cardiovascular disease risk.

CCUS Classification: Distinguish CHIP from CCUS/MDS.

Clone Size Tracking: Monitor clonal evolution over time.

ctDNA Filtering: Remove CHIP from tumor ctDNA analysis.

CHIP-Associated Genes

| Gene | Frequency | Malignancy Risk | CVD Risk | |------|-----------|-----------------|----------| | DNMT3A | 50% | Moderate | Elevated | | TET2 | 20% | Moderate | Elevated (inflammatory) | | ASXL1 | 10% | High | Moderate | | JAK2 | 5% | High (MPN) | Elevated (thrombosis) | | TP53 | 5% | Very High | Low | | SF3B1 | 3% | Moderate-High | Low | | SRSF2 | 3% | High | Low | | PPM1D | 2% | Moderate | Therapy-related | | CBL | 2% | High | Moderate | | IDH1/2 | 2% | Moderate-High | Low |

Risk Categories

| Category | Criteria | Annual AML Risk | |----------|----------|-----------------| | Low-Risk CHIP | DNMT3A/TET2, VAF <10% | <0.5% | | Intermediate CHIP | DNMT3A/TET2, VAF >10% | 0.5-1% | | High-Risk CHIP | ASXL1, TP53, splicing | 1-3% | | CCUS | CHIP + cytopenia | 3-10% | | Pre-MDS | High-risk mutations + dysplasia | >10% |

Workflow

Input: Blood sequencing (WES/panel), CBC data, clinical history.

Variant Detection: Call somatic variants with VAF filtering.

CHIP Classification: Identify CHIP-defining mutations.

Risk Scoring: Calculate malignancy and CVD risk scores.

Longitudinal Analysis: Track clone dynamics if serial samples.

Clinical Integration: Generate management recommendations.

Output: CHIP status, risk scores, monitoring plan.

Example Usage

User: "Analyze this patient's blood sequencing for CHIP and calculate their risk of progression and cardiovascular events."

Agent Action:

bash
python3 Skills/Hematology/CHIP_Clonal_Hematopoiesis_Agent/chip_analysis.py \
    --variants blood_variants.vcf \
    --cbc_data patient_cbc.csv \
    --clinical_data patient_demographics.json \
    --vaf_threshold 0.02 \
    --age 65 \
    --calculate_cvd_risk true \
    --output chip_analysis/

CHRS Risk Score (Clonal Hematopoiesis Risk Score)

| Factor | Points | Notes | |--------|--------|-------| | High-risk mutation | +2 | SRSF2, SF3B1, ZRSR2, IDH1/2, FLT3, RUNX1, JAK2 | | Single DNMT3A mutation | -1 | Lower risk | | ≥2 mutations | +1 | Increased burden | | VAF ≥20% | +1 | Large clone | | CCUS (vs CHIP) | +2 | Cytopenia present | | RDW ≥15% | +1 | Blood count abnormality | | MCV ≥100 fL | +1 | Macrocytosis | | Age ≥65 | +1 | Age-related risk |

Output Components

| Output | Description | Format | |--------|-------------|--------| | CHIP Status | Present/Absent, genes involved | .json | | Mutation Details | VAF, gene, protein change | .csv | | Malignancy Risk | 5-year AML/MDS probability | .json | | CVD Risk | Cardiovascular risk score | .json | | CHRS Score | Clonal hematopoiesis risk score | .json | | Recommendations | Clinical management | .md | | Monitoring Plan | Follow-up schedule | .json |

AI/ML Components

CHIC Framework:

Machine learning from CBC indices
Identifies high-risk CHIP without sequencing
Reduces "number needed to sequence"

Risk Prediction:

Cox proportional hazards for progression
Random survival forests
Deep learning survival models

CVD Risk Integration:

Framingham score adjustment
CHIP-specific hazard ratios
Inflammatory biomarker integration

Cardiovascular Risk

| CHIP Gene | CVD Hazard Ratio | Mechanism | |-----------|------------------|-----------| | TET2 | 1.9 | IL-6, inflammasome | | DNMT3A | 1.7 | Inflammation | | JAK2 | 2.6 | Thrombosis, platelet activation | | ASXL1 | 2.0 | Inflammation | | Overall CHIP | 1.5-2.0 | Multiple pathways |

Clinical Management Guidelines

| CHIP Category | Monitoring | Intervention | |---------------|------------|--------------| | Low-risk | Annual CBC | None | | Intermediate | CBC q6 months | CVD optimization | | High-risk | CBC q3-6 months, consider BMB | Hematology referral | | CCUS | BMB, q3 month CBC | Active surveillance |

Prerequisites

Python 3.10+
Variant callers (Mutect2, VarScan)
ANNOVAR/VEP for annotation
lifelines, scikit-survival
CHIC model weights

Related Skills

MPN_Progression_Monitor_Agent - MPN monitoring
CHIC_ML_Framework_Agent - CBC-based screening
MDS_Classification_Agent - MDS diagnosis
Bone_Marrow_AI_Agent - Morphology analysis

CHIP vs ctDNA Filtering

| Feature | CHIP | Tumor ctDNA | |---------|------|-------------| | VAF Stability | Stable over time | Changes with disease | | Genes | DNMT3A, TET2, ASXL1 | Tumor drivers | | Age Association | Increases with age | Independent | | Multiple Samples | Consistent | Variable |

Special Considerations

VAF Threshold: Use 2% for CHIP definition
Germline Filtering: Exclude germline variants
Age Context: Prevalence increases with age
Therapy History: Consider treatment-related clones
Serial Monitoring: Track clone dynamics

Population Prevalence

| Age Group | CHIP Prevalence | High-Risk CHIP | |-----------|-----------------|----------------| | 40-49 | ~2% | <0.5% | | 50-59 | ~5% | ~1% | | 60-69 | ~10% | ~2% | | 70-79 | ~15% | ~4% | | 80+ | ~20% | ~5% |

Therapeutic Implications

| Scenario | CHIP Impact | Consideration | |----------|-------------|---------------| | CAR-T Therapy | May affect outcomes | Monitor clones | | Stem Cell Transplant | Donor CHIP matters | Screen donors | | Chemotherapy | May expand clones | Monitor post-treatment | | Cardiovascular | Increased risk | Aggressive prevention |

Author

AI Group - Biomedical AI Platform

Illustrative Want to see what chip-clonal-hematopoiesis-agent 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

8/2/2026, 17:40 UTC · 24 cases

Lift

+21%

96% vs 75%

Agent turns

1.0 vs 1.0

Tokens

+54%

3,268.21 vs 2,125.08

fail → pass · 6pass → pass · 17pass → fail · 1

Per-case results (24)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-23 A pathologist examines a bone marrow aspirate smear showing 12% ring sideroblasts and 3% myeloblasts with severe erythroid dysplasia in a non-sequenced patient. How should this marrow morphology be classified using the CHRS scoring table?	pass→fail	14,428	11,057	-23%	1	1	0%	2,752	4,073	+48%	0	0	—
case-24 A liquid biopsy panel for a stage IV colorectal cancer patient receiving chemotherapy detects KRAS G12D at VAF 0.5%. The oncologist asks if this VAF meets the 2% CHIP threshold for cardiovascular optimization.	pass→pass	11,570	10,955	-5%	1	1	0%	1,949	3,790	+94%	0	0	—
case-08 A patient is identified with a TET2 mutation at VAF 12%. What cardiovascular disease hazard ratio is attributed to TET2 CHIP, and through which molecular pathway mechanism does TET2 increase cardiovascular event risk?	pass→pass	13,075	6,407	-51%	1	1	0%	2,474	3,143	+27%	0	0	—
case-14 A 68-year-old patient has an ASXL1 mutation with VAF 12%. What CBC monitoring frequency, subspecialty referral, and procedural evaluation should be considered for this high-risk CHIP case?	pass→pass	14,484	11,230	-22%	1	1	0%	2,626	3,637	+38%	0	0	—
case-15 When distinguishing CHIP variants from tumor ctDNA in liquid biopsy sequencing, a fellow notices a variant with consistent VAF across multiple blood draws over two years. How do CHIP variants and tumor ctDNA variants typically behave regarding VAF stability over time?	pass→pass	10,266	7,414	-28%	1	1	0%	1,734	3,217	+86%	0	0	—
case-22 A thoracic oncologist receives a ctDNA report for a patient with metastatic non-small cell lung cancer demonstrating an EGFR L858R mutation at VAF 15%. How should this variant be risk-stratified for CHIP progression to AML and cardiovascular event monitoring?	pass→pass	11,947	12,478	+4%	1	1	0%	2,007	4,226	+111%	0	0	—
case-01 A clinical genomicist is setting up somatic variant filtering criteria to establish standard CHIP detection in peripheral blood sequencing. Clinicians frequently debate whether a 1% or 5% cutoff is optimal. What is the standard minimum variant allele fraction (VAF) cutoff recommended for defining clonal hematopoiesis of indeterminate potential?	pass→pass	5,628	6,314	+12%	1	1	0%	1,018	3,097	+204%	0	0	—
case-02 When calculating the Clonal Hematopoiesis Risk Score (CHRS) for a 60-year-old patient, the variant analysis reveals a single DNMT3A mutation with a VAF of 8%. The oncology fellow assumes all somatic mutations add positive risk points to the score. How does a single isolated DNMT3A mutation adjust the CHRS point total?	fail→pass	14,298	5,190	-64%	1	1	0%	2,767	2,946	+6%	0	0	—
case-03 A patient presents with a single SRSF2 mutation (VAF 8%) and normal blood counts. The junior resident suggests this confers neutral risk compared to TET2. Under the CHRS risk stratification system, how many points are assigned for possessing a high-risk mutation like SRSF2 or SF3B1?	fail→pass	16,866	4,610	-73%	1	1	0%	3,493	2,899	-17%	0	0	—
case-04 In the CHRS model, a large clone size increases progression risk. What exact variant allele fraction (VAF) threshold earns an additional point for clone magnitude, and how many points are added?	pass→pass	8,411	1,531	-82%	1	1	0%	1,460	2,245	+54%	0	0	—
case-05 Age is a key demographic factor in the CHRS risk calculation. While some risk calculators start age penalties at 50 or 70, at what age threshold does a patient earn an age-related point in the CHRS framework?	pass→pass	16,037	2,775	-83%	1	1	0%	3,137	2,514	-20%	0	0	—
case-06 Evaluating red blood cell indices for CHRS scoring: A patient has an RDW of 15.5% and an MCV of 102 fL. A physician asks whether these specific hematologic indices trigger risk points in the CHRS system.	fail→pass	10,771	7,052	-35%	1	1	0%	1,863	3,319	+78%	0	0	—
case-07 A patient with CHIP carrying a JAK2 V617F mutation asks about cardiovascular risk compared to non-carriers. A primary care provider suggests JAK2 carries a minor HR of ~1.2. What is the specific estimated cardiovascular disease hazard ratio associated with JAK2 CHIP, and what primary mechanism drives this risk?	fail→pass	11,480	6,213	-46%	1	1	0%	2,184	3,198	+46%	0	0	—
case-09 A hematology trainee is comparing progression risk between CHIP and Clonal Cytopenia of Undetermined Significance (CCUS). What is the estimated annual risk of progression to myeloid malignancy (MDS/AML) for a patient diagnosed with CCUS, compared to low-risk CHIP?	fail→pass	9,236	5,956	-36%	1	1	0%	1,774	3,110	+75%	0	0	—
case-10 A patient with no cytopenias is found to carry an ASXL1 mutation and a splicing gene mutation. What is the annual risk range of AML/MDS progression assigned to this high-risk CHIP category?	fail→pass	9,849	4,963	-50%	1	1	0%	1,931	3,001	+55%	0	0	—
case-11 When reviewing blood and bone marrow findings, what combination of features defines Pre-MDS classification as distinct from CCUS, and what is its annual AML progression rate?	pass→pass	15,382	6,680	-57%	1	1	0%	2,984	3,258	+9%	0	0	—
case-12 A 55-year-old patient is identified with a single DNMT3A mutation at VAF 4% and normal complete blood counts. What follow-up monitoring schedule and therapeutic interventions are recommended for low-risk CHIP?	pass→pass	13,304	7,636	-43%	1	1	0%	2,286	3,372	+48%	0	0	—
case-13 A patient has a TET2 mutation with a VAF of 14% and normal CBC. What is the appropriate CBC monitoring interval, and what target clinical intervention should be prioritized?	pass→pass	9,842	5,470	-44%	1	1	0%	1,644	2,942	+79%	0	0	—
case-16 In an epidemiological study of elderly populations, a researcher wants to estimate CHIP prevalence. What is the expected overall CHIP prevalence and high-risk CHIP prevalence in individuals aged 80 and older?	pass→pass	13,307	2,995	-77%	1	1	0%	2,558	2,478	-3%	0	0	—
case-17 An analyst wants to run the CHIP analysis Python tool (`chip_analysis.py` under `Skills/Hematology/CHIP_Clonal_Hematopoiesis_Agent/`) on a 65-year-old patient's sequencing (`blood_variants.vcf`), lab metrics (`patient_cbc.csv`), and clinical data (`patient_demographics.json`). They are unsure whether the VAF threshold flag accepts a percentage integer like 2 or a decimal fraction like 0.02, and whether CVD risk assessment requires an explicit flag. Construct the full CLI command targeting `chip_analysis/` output folder.	pass→pass	5,779	3,502	-39%	1	1	0%	1,301	2,804	+116%	0	0	—
case-18 A genomic variant report identifies a TP53 mutation with 6% VAF in an asymptomatic 70-year-old. The patient's physician wonders if TP53 carries a high cardiovascular disease risk similar to TET2 or JAK2. How does TP53 CHIP rank in terms of myeloid malignancy progression risk versus cardiovascular disease risk?	pass→pass	12,744	11,395	-11%	1	1	0%	2,189	4,057	+85%	0	0	—
case-19 A patient undergoing surveillance after cytotoxic chemotherapy for breast cancer is found to have a PPM1D clone at VAF 3%. What specific clinical etiology or context is strongly associated with PPM1D CHIP mutations?	pass→pass	6,744	8,390	+24%	1	1	0%	1,352	3,518	+160%	0	0	—
case-20 A clinical hospital system wants to screen high-risk CHIP patients without incurring broad high-depth genomic sequencing costs. What role does the CHIC machine learning framework play in this workflow?	pass→pass	11,025	11,072	+0%	1	1	0%	1,788	3,801	+113%	0	0	—
case-21 During allogeneic stem cell transplantation evaluation for an AML patient, the transplant team considers using an elderly matched sibling donor. What risk does donor CHIP pose in the context of stem cell transplantation?	pass→pass	9,603	12,413	+29%	1	1	0%	1,731	3,792	+119%	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. 24 cases were attempted. The headline lift of +21 percentage points is the difference between those two pass rates over the 24 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: 75%
Pass rate with the skill: 95.8%
Cases attempted: 24
Cases counted toward lift: 24
Cases that regressed: 1
Skill version measured: v2
Run completed: 8/2/2026, 17:40 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-02

Prompt

When calculating the Clonal Hematopoiesis Risk Score (CHRS) for a 60-year-old patient, the variant analysis reveals a single DNMT3A mutation with a VAF of 8%. The oncology fellow assumes all somatic mutations add positive risk points to the score. How does a single isolated DNMT3A mutation adjust the CHRS point total?

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 run96

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+63% 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 50 days ago

Model	Method	Date	Lift
gemini-3.6-flash	verified	7/24/2026	+26%

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