[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-nvidia-analysis-methods":3,"mdc-juyfct-key":37,"related-org-nvidia-analysis-methods":989,"related-repo-nvidia-analysis-methods":1148},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":26,"repoUrl":27,"updatedAt":28,"license":29,"forks":30,"topics":31,"repo":32,"sourceUrl":35,"mdContent":36},"analysis-methods","write Python analysis code for FHIR data","Teaches the analyst agent how to write correct, robust Python analysis code for FHIR clinical data using pandas, matplotlib, and scipy.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},"nvidia","NVIDIA","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fnvidia.png",[12,16,19,20,23],{"name":13,"slug":14,"type":15},"Healthcare","healthcare","tag",{"name":17,"slug":18,"type":15},"Data Analysis","data-analysis",{"name":9,"slug":8,"type":15},{"name":21,"slug":22,"type":15},"Python","python",{"name":24,"slug":25,"type":15},"FHIR","fhir",1144,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fdgx-spark-playbooks","2026-07-14T05:35:59.037962",null,249,[],{"repoUrl":27,"stars":26,"forks":30,"topics":33,"description":34},[],"Collection of step-by-step playbooks for setting up AI\u002FML workloads on NVIDIA DGX Spark devices with Blackwell architecture.","https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fdgx-spark-playbooks\u002Ftree\u002FHEAD\u002Fnvidia\u002Fstation-healthcare-agent\u002Fassets\u002Fskills\u002Fanalysis-methods","---\nname: analysis-methods\ndescription: Teaches the analyst agent how to write correct, robust Python analysis code for FHIR clinical data using pandas, matplotlib, and scipy.\nmetadata:\n  openclaw:\n    requires:\n      bins: [\"python3\", \"pip\"]\n---\n\n# Analysis Code Guidelines\n\n## FHIR Helpers Library\n\n**Always import the helpers library at the top of every analysis script:**\n\n```python\nimport sys\nsys.path.insert(0, '\u002Fsandbox\u002Fclinical-intelligence\u002Fskills\u002Fanalysis-methods\u002Fscripts')\nfrom fhir_helpers import *\n```\n\n### Available functions\n\n| Function | Use for | HTTP calls |\n|----------|---------|------------|\n| `get_patients_with_condition(snomed_code)` | Find patients with a condition → list of IDs | 1-2 |\n| `get_latest_labs_batch(loinc_code, patient_ids)` | Labs for a cohort → dict: pid → (value, unit, date) | 1-2 |\n| `get_all_medications_batch(patient_ids)` | Meds for a cohort → dict: pid → [med names] | 1-2 |\n| `build_cohort_df(patient_ids, loinc, lab_name, drug_check_fn)` | Full DataFrame with labs + meds | 2-3 |\n| `get_latest_lab(patient_id, loinc_code)` | Lab for ONE patient → (value, unit, date) | 1 |\n| `get_medications(patient_id)` | Meds for ONE patient → [names] | 1 |\n| `get_latest_bp(patient_id)` | BP for ONE patient → (sys, dia, date) | 1-2 |\n| `check_drug_class(med_list, drug_names)` | Check if any med matches drug list → bool | 0 |\n| `fhir_get(path, params)` | Raw FHIR GET → parsed JSON | 1 |\n| `get_all_pages(path, params)` | Paginated FHIR GET → all entries | 1+ |\n| `save_chart_to_canvas(fig, filename)` | Save matplotlib figure to canvas directory | 0 |\n\n### Performance rules\n\n- **Cohort queries (2+ patients):** Use `get_latest_labs_batch()` and `get_all_medications_batch()`. These make 1-2 HTTP calls total regardless of patient count.\n- **Single patient:** Use `get_latest_lab()`, `get_medications()`, `get_latest_bp()`.\n- **NEVER** loop over patients calling `get_latest_lab()` per patient. Each HTTP call through the sandbox proxy adds 1-3s. For 48 patients = 48 calls = 2+ minutes. The batch function does it in one call.\n\n## Execution Rules\n\n- Run scripts with `python` (NOT `python3`)\n- Write a SINGLE Python script for the entire task\n- Write the script to `\u002Ftmp\u002F\u003Cname>.py`, then execute it\n- All HTTP inside the sandbox must use `subprocess.run([\"curl\", ...])` — the `requests` library does NOT work\n\n## Mandatory Workflow\n\n```\nSTEP 1 - WRITE SCRIPT (import fhir_helpers, write analysis)\nSTEP 2 - VALIDATE: python \u002Fsandbox\u002Fclinical-intelligence\u002Fscripts\u002Fvalidate_and_run.py --validate-only \u002Ftmp\u002F\u003Cname>.py\nSTEP 3 - EXECUTE: python \u002Ftmp\u002F\u003Cname>.py\nSTEP 4 - INTERPRET: explain results using clinical-knowledge skill\n```\n\n## Code Structure\n\n1. Imports (always start with fhir_helpers import)\n2. Data collection (use batch functions)\n3. DataFrame construction\n4. Analysis (filters, aggregations)\n5. Visualization -- use `save_chart_to_canvas(fig, filename)` (NOT plt.savefig)\n6. Summary (print findings)\n7. Disclaimer\n\n## Care Gap Analysis Pattern\n\n```python\n# Example: diabetes care gap\npatients = get_patients_with_condition(\"44054006\")  # SNOMED for diabetes\ndf = build_cohort_df(patients, \"4548-4\", \"HbA1c\",\n                     lambda meds: check_drug_class(meds, [\"metformin\", \"insulin\", \"glipizide\"]))\n\ngap = df[(df['HbA1c'] > 9) & (~df['on_target_med'])]\ndenom = len(df[df['HbA1c'].notna()])\npct = f\"{len(gap)\u002Fdenom*100:.1f}%\" if denom > 0 else \"N\u002FA (no HbA1c data)\"\nprint(f\"Care gap: {len(gap)}\u002F{denom} ({pct})\")\n```\n\n## Visualization\n\nAlways use dark theme. Use `save_chart_to_canvas()` instead of `plt.savefig()` directly.\n\n```python\nimport matplotlib\nmatplotlib.use('Agg')\nimport matplotlib.pyplot as plt\n\nplt.style.use('dark_background')\nfig, ax = plt.subplots(figsize=(10, 6))\nfig.patch.set_facecolor('#1a1a1a')\nax.set_facecolor('#1a1a1a')\n\n# Histogram with NVIDIA green\nax.hist(values, bins=15, color='#76B900', edgecolor='#1a1a1a', alpha=0.85)\nax.axvline(x=threshold, color='#ff4444', linestyle='--', linewidth=2, label=f'Threshold ({threshold})')\nax.set_title(\"Title\", fontsize=14, fontweight='bold', color='white')\nax.legend()\nax.grid(axis='y', alpha=0.2, color='#444444')\nax.text(0.98, 0.95, f\"N = {len(values)}\", transform=ax.transAxes, fontsize=11, color='#888888', ha='right', va='top')\n\n# MANDATORY: use save_chart_to_canvas (NOT plt.savefig)\nsave_chart_to_canvas(fig, \"chart.png\")\nplt.close()\n```\n\n## Guardrails\n\n- Never compute statistics on fewer than 5 data points\n- Always report sample size: \"45.0% (27 out of 60)\"\n- Flag data quality issues if >30% missing\n- Do not fabricate data — report what exists, flag what's missing\n- All charts must include N annotation\n\n## Output Format\n\nEnd every script with:\n\n```python\nprint(f\"\\nDisclaimer: This analysis is for research and operational purposes.\")\nprint(\"Clinical decisions should be made by qualified clinicians.\")\n```\n",{"data":38,"body":45},{"name":4,"description":6,"metadata":39},{"openclaw":40},{"requires":41},{"bins":42},[43,44],"python3","pip",{"type":46,"children":47},"root",[48,57,64,74,114,121,400,406,486,492,553,559,569,575,621,627,713,719,740,915,921,949,955,960,983],{"type":49,"tag":50,"props":51,"children":53},"element","h1",{"id":52},"analysis-code-guidelines",[54],{"type":55,"value":56},"text","Analysis Code Guidelines",{"type":49,"tag":58,"props":59,"children":61},"h2",{"id":60},"fhir-helpers-library",[62],{"type":55,"value":63},"FHIR Helpers Library",{"type":49,"tag":65,"props":66,"children":67},"p",{},[68],{"type":49,"tag":69,"props":70,"children":71},"strong",{},[72],{"type":55,"value":73},"Always import the helpers library at the top of every analysis script:",{"type":49,"tag":75,"props":76,"children":80},"pre",{"className":77,"code":78,"language":22,"meta":79,"style":79},"language-python shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","import sys\nsys.path.insert(0, '\u002Fsandbox\u002Fclinical-intelligence\u002Fskills\u002Fanalysis-methods\u002Fscripts')\nfrom fhir_helpers import *\n","",[81],{"type":49,"tag":82,"props":83,"children":84},"code",{"__ignoreMap":79},[85,96,105],{"type":49,"tag":86,"props":87,"children":90},"span",{"class":88,"line":89},"line",1,[91],{"type":49,"tag":86,"props":92,"children":93},{},[94],{"type":55,"value":95},"import sys\n",{"type":49,"tag":86,"props":97,"children":99},{"class":88,"line":98},2,[100],{"type":49,"tag":86,"props":101,"children":102},{},[103],{"type":55,"value":104},"sys.path.insert(0, '\u002Fsandbox\u002Fclinical-intelligence\u002Fskills\u002Fanalysis-methods\u002Fscripts')\n",{"type":49,"tag":86,"props":106,"children":108},{"class":88,"line":107},3,[109],{"type":49,"tag":86,"props":110,"children":111},{},[112],{"type":55,"value":113},"from fhir_helpers import *\n",{"type":49,"tag":115,"props":116,"children":118},"h3",{"id":117},"available-functions",[119],{"type":55,"value":120},"Available functions",{"type":49,"tag":122,"props":123,"children":124},"table",{},[125,149],{"type":49,"tag":126,"props":127,"children":128},"thead",{},[129],{"type":49,"tag":130,"props":131,"children":132},"tr",{},[133,139,144],{"type":49,"tag":134,"props":135,"children":136},"th",{},[137],{"type":55,"value":138},"Function",{"type":49,"tag":134,"props":140,"children":141},{},[142],{"type":55,"value":143},"Use for",{"type":49,"tag":134,"props":145,"children":146},{},[147],{"type":55,"value":148},"HTTP calls",{"type":49,"tag":150,"props":151,"children":152},"tbody",{},[153,176,197,223,245,267,293,314,336,357,379],{"type":49,"tag":130,"props":154,"children":155},{},[156,166,171],{"type":49,"tag":157,"props":158,"children":159},"td",{},[160],{"type":49,"tag":82,"props":161,"children":163},{"className":162},[],[164],{"type":55,"value":165},"get_patients_with_condition(snomed_code)",{"type":49,"tag":157,"props":167,"children":168},{},[169],{"type":55,"value":170},"Find patients with a condition → list of IDs",{"type":49,"tag":157,"props":172,"children":173},{},[174],{"type":55,"value":175},"1-2",{"type":49,"tag":130,"props":177,"children":178},{},[179,188,193],{"type":49,"tag":157,"props":180,"children":181},{},[182],{"type":49,"tag":82,"props":183,"children":185},{"className":184},[],[186],{"type":55,"value":187},"get_latest_labs_batch(loinc_code, patient_ids)",{"type":49,"tag":157,"props":189,"children":190},{},[191],{"type":55,"value":192},"Labs for a cohort → dict: pid → (value, unit, date)",{"type":49,"tag":157,"props":194,"children":195},{},[196],{"type":55,"value":175},{"type":49,"tag":130,"props":198,"children":199},{},[200,209,219],{"type":49,"tag":157,"props":201,"children":202},{},[203],{"type":49,"tag":82,"props":204,"children":206},{"className":205},[],[207],{"type":55,"value":208},"get_all_medications_batch(patient_ids)",{"type":49,"tag":157,"props":210,"children":211},{},[212,214],{"type":55,"value":213},"Meds for a cohort → dict: pid → ",{"type":49,"tag":86,"props":215,"children":216},{},[217],{"type":55,"value":218},"med names",{"type":49,"tag":157,"props":220,"children":221},{},[222],{"type":55,"value":175},{"type":49,"tag":130,"props":224,"children":225},{},[226,235,240],{"type":49,"tag":157,"props":227,"children":228},{},[229],{"type":49,"tag":82,"props":230,"children":232},{"className":231},[],[233],{"type":55,"value":234},"build_cohort_df(patient_ids, loinc, lab_name, drug_check_fn)",{"type":49,"tag":157,"props":236,"children":237},{},[238],{"type":55,"value":239},"Full DataFrame with labs + meds",{"type":49,"tag":157,"props":241,"children":242},{},[243],{"type":55,"value":244},"2-3",{"type":49,"tag":130,"props":246,"children":247},{},[248,257,262],{"type":49,"tag":157,"props":249,"children":250},{},[251],{"type":49,"tag":82,"props":252,"children":254},{"className":253},[],[255],{"type":55,"value":256},"get_latest_lab(patient_id, loinc_code)",{"type":49,"tag":157,"props":258,"children":259},{},[260],{"type":55,"value":261},"Lab for ONE patient → (value, unit, date)",{"type":49,"tag":157,"props":263,"children":264},{},[265],{"type":55,"value":266},"1",{"type":49,"tag":130,"props":268,"children":269},{},[270,279,289],{"type":49,"tag":157,"props":271,"children":272},{},[273],{"type":49,"tag":82,"props":274,"children":276},{"className":275},[],[277],{"type":55,"value":278},"get_medications(patient_id)",{"type":49,"tag":157,"props":280,"children":281},{},[282,284],{"type":55,"value":283},"Meds for ONE patient → ",{"type":49,"tag":86,"props":285,"children":286},{},[287],{"type":55,"value":288},"names",{"type":49,"tag":157,"props":290,"children":291},{},[292],{"type":55,"value":266},{"type":49,"tag":130,"props":294,"children":295},{},[296,305,310],{"type":49,"tag":157,"props":297,"children":298},{},[299],{"type":49,"tag":82,"props":300,"children":302},{"className":301},[],[303],{"type":55,"value":304},"get_latest_bp(patient_id)",{"type":49,"tag":157,"props":306,"children":307},{},[308],{"type":55,"value":309},"BP for ONE patient → (sys, dia, date)",{"type":49,"tag":157,"props":311,"children":312},{},[313],{"type":55,"value":175},{"type":49,"tag":130,"props":315,"children":316},{},[317,326,331],{"type":49,"tag":157,"props":318,"children":319},{},[320],{"type":49,"tag":82,"props":321,"children":323},{"className":322},[],[324],{"type":55,"value":325},"check_drug_class(med_list, drug_names)",{"type":49,"tag":157,"props":327,"children":328},{},[329],{"type":55,"value":330},"Check if any med matches drug list → bool",{"type":49,"tag":157,"props":332,"children":333},{},[334],{"type":55,"value":335},"0",{"type":49,"tag":130,"props":337,"children":338},{},[339,348,353],{"type":49,"tag":157,"props":340,"children":341},{},[342],{"type":49,"tag":82,"props":343,"children":345},{"className":344},[],[346],{"type":55,"value":347},"fhir_get(path, params)",{"type":49,"tag":157,"props":349,"children":350},{},[351],{"type":55,"value":352},"Raw FHIR GET → parsed JSON",{"type":49,"tag":157,"props":354,"children":355},{},[356],{"type":55,"value":266},{"type":49,"tag":130,"props":358,"children":359},{},[360,369,374],{"type":49,"tag":157,"props":361,"children":362},{},[363],{"type":49,"tag":82,"props":364,"children":366},{"className":365},[],[367],{"type":55,"value":368},"get_all_pages(path, params)",{"type":49,"tag":157,"props":370,"children":371},{},[372],{"type":55,"value":373},"Paginated FHIR GET → all entries",{"type":49,"tag":157,"props":375,"children":376},{},[377],{"type":55,"value":378},"1+",{"type":49,"tag":130,"props":380,"children":381},{},[382,391,396],{"type":49,"tag":157,"props":383,"children":384},{},[385],{"type":49,"tag":82,"props":386,"children":388},{"className":387},[],[389],{"type":55,"value":390},"save_chart_to_canvas(fig, filename)",{"type":49,"tag":157,"props":392,"children":393},{},[394],{"type":55,"value":395},"Save matplotlib figure to canvas directory",{"type":49,"tag":157,"props":397,"children":398},{},[399],{"type":55,"value":335},{"type":49,"tag":115,"props":401,"children":403},{"id":402},"performance-rules",[404],{"type":55,"value":405},"Performance rules",{"type":49,"tag":407,"props":408,"children":409},"ul",{},[410,437,469],{"type":49,"tag":411,"props":412,"children":413},"li",{},[414,419,421,427,429,435],{"type":49,"tag":69,"props":415,"children":416},{},[417],{"type":55,"value":418},"Cohort queries (2+ patients):",{"type":55,"value":420}," Use ",{"type":49,"tag":82,"props":422,"children":424},{"className":423},[],[425],{"type":55,"value":426},"get_latest_labs_batch()",{"type":55,"value":428}," and ",{"type":49,"tag":82,"props":430,"children":432},{"className":431},[],[433],{"type":55,"value":434},"get_all_medications_batch()",{"type":55,"value":436},". These make 1-2 HTTP calls total regardless of patient count.",{"type":49,"tag":411,"props":438,"children":439},{},[440,445,446,452,454,460,461,467],{"type":49,"tag":69,"props":441,"children":442},{},[443],{"type":55,"value":444},"Single patient:",{"type":55,"value":420},{"type":49,"tag":82,"props":447,"children":449},{"className":448},[],[450],{"type":55,"value":451},"get_latest_lab()",{"type":55,"value":453},", ",{"type":49,"tag":82,"props":455,"children":457},{"className":456},[],[458],{"type":55,"value":459},"get_medications()",{"type":55,"value":453},{"type":49,"tag":82,"props":462,"children":464},{"className":463},[],[465],{"type":55,"value":466},"get_latest_bp()",{"type":55,"value":468},".",{"type":49,"tag":411,"props":470,"children":471},{},[472,477,479,484],{"type":49,"tag":69,"props":473,"children":474},{},[475],{"type":55,"value":476},"NEVER",{"type":55,"value":478}," loop over patients calling ",{"type":49,"tag":82,"props":480,"children":482},{"className":481},[],[483],{"type":55,"value":451},{"type":55,"value":485}," per patient. Each HTTP call through the sandbox proxy adds 1-3s. For 48 patients = 48 calls = 2+ minutes. The batch function does it in one call.",{"type":49,"tag":58,"props":487,"children":489},{"id":488},"execution-rules",[490],{"type":55,"value":491},"Execution Rules",{"type":49,"tag":407,"props":493,"children":494},{},[495,514,519,532],{"type":49,"tag":411,"props":496,"children":497},{},[498,500,505,507,512],{"type":55,"value":499},"Run scripts with ",{"type":49,"tag":82,"props":501,"children":503},{"className":502},[],[504],{"type":55,"value":22},{"type":55,"value":506}," (NOT ",{"type":49,"tag":82,"props":508,"children":510},{"className":509},[],[511],{"type":55,"value":43},{"type":55,"value":513},")",{"type":49,"tag":411,"props":515,"children":516},{},[517],{"type":55,"value":518},"Write a SINGLE Python script for the entire task",{"type":49,"tag":411,"props":520,"children":521},{},[522,524,530],{"type":55,"value":523},"Write the script to ",{"type":49,"tag":82,"props":525,"children":527},{"className":526},[],[528],{"type":55,"value":529},"\u002Ftmp\u002F\u003Cname>.py",{"type":55,"value":531},", then execute it",{"type":49,"tag":411,"props":533,"children":534},{},[535,537,543,545,551],{"type":55,"value":536},"All HTTP inside the sandbox must use ",{"type":49,"tag":82,"props":538,"children":540},{"className":539},[],[541],{"type":55,"value":542},"subprocess.run([\"curl\", ...])",{"type":55,"value":544}," — the ",{"type":49,"tag":82,"props":546,"children":548},{"className":547},[],[549],{"type":55,"value":550},"requests",{"type":55,"value":552}," library does NOT work",{"type":49,"tag":58,"props":554,"children":556},{"id":555},"mandatory-workflow",[557],{"type":55,"value":558},"Mandatory Workflow",{"type":49,"tag":75,"props":560,"children":564},{"className":561,"code":563,"language":55},[562],"language-text","STEP 1 - WRITE SCRIPT (import fhir_helpers, write analysis)\nSTEP 2 - VALIDATE: python \u002Fsandbox\u002Fclinical-intelligence\u002Fscripts\u002Fvalidate_and_run.py --validate-only \u002Ftmp\u002F\u003Cname>.py\nSTEP 3 - EXECUTE: python \u002Ftmp\u002F\u003Cname>.py\nSTEP 4 - INTERPRET: explain results using clinical-knowledge skill\n",[565],{"type":49,"tag":82,"props":566,"children":567},{"__ignoreMap":79},[568],{"type":55,"value":563},{"type":49,"tag":58,"props":570,"children":572},{"id":571},"code-structure",[573],{"type":55,"value":574},"Code Structure",{"type":49,"tag":576,"props":577,"children":578},"ol",{},[579,584,589,594,599,611,616],{"type":49,"tag":411,"props":580,"children":581},{},[582],{"type":55,"value":583},"Imports (always start with fhir_helpers import)",{"type":49,"tag":411,"props":585,"children":586},{},[587],{"type":55,"value":588},"Data collection (use batch functions)",{"type":49,"tag":411,"props":590,"children":591},{},[592],{"type":55,"value":593},"DataFrame construction",{"type":49,"tag":411,"props":595,"children":596},{},[597],{"type":55,"value":598},"Analysis (filters, aggregations)",{"type":49,"tag":411,"props":600,"children":601},{},[602,604,609],{"type":55,"value":603},"Visualization -- use ",{"type":49,"tag":82,"props":605,"children":607},{"className":606},[],[608],{"type":55,"value":390},{"type":55,"value":610}," (NOT plt.savefig)",{"type":49,"tag":411,"props":612,"children":613},{},[614],{"type":55,"value":615},"Summary (print findings)",{"type":49,"tag":411,"props":617,"children":618},{},[619],{"type":55,"value":620},"Disclaimer",{"type":49,"tag":58,"props":622,"children":624},{"id":623},"care-gap-analysis-pattern",[625],{"type":55,"value":626},"Care Gap Analysis Pattern",{"type":49,"tag":75,"props":628,"children":630},{"className":77,"code":629,"language":22,"meta":79,"style":79},"# Example: diabetes care gap\npatients = get_patients_with_condition(\"44054006\")  # SNOMED for diabetes\ndf = build_cohort_df(patients, \"4548-4\", \"HbA1c\",\n                     lambda meds: check_drug_class(meds, [\"metformin\", \"insulin\", \"glipizide\"]))\n\ngap = df[(df['HbA1c'] > 9) & (~df['on_target_med'])]\ndenom = len(df[df['HbA1c'].notna()])\npct = f\"{len(gap)\u002Fdenom*100:.1f}%\" if denom > 0 else \"N\u002FA (no HbA1c data)\"\nprint(f\"Care gap: {len(gap)}\u002F{denom} ({pct})\")\n",[631],{"type":49,"tag":82,"props":632,"children":633},{"__ignoreMap":79},[634,642,650,658,667,677,686,695,704],{"type":49,"tag":86,"props":635,"children":636},{"class":88,"line":89},[637],{"type":49,"tag":86,"props":638,"children":639},{},[640],{"type":55,"value":641},"# Example: diabetes care gap\n",{"type":49,"tag":86,"props":643,"children":644},{"class":88,"line":98},[645],{"type":49,"tag":86,"props":646,"children":647},{},[648],{"type":55,"value":649},"patients = get_patients_with_condition(\"44054006\")  # SNOMED for diabetes\n",{"type":49,"tag":86,"props":651,"children":652},{"class":88,"line":107},[653],{"type":49,"tag":86,"props":654,"children":655},{},[656],{"type":55,"value":657},"df = build_cohort_df(patients, \"4548-4\", \"HbA1c\",\n",{"type":49,"tag":86,"props":659,"children":661},{"class":88,"line":660},4,[662],{"type":49,"tag":86,"props":663,"children":664},{},[665],{"type":55,"value":666},"                     lambda meds: check_drug_class(meds, [\"metformin\", \"insulin\", \"glipizide\"]))\n",{"type":49,"tag":86,"props":668,"children":670},{"class":88,"line":669},5,[671],{"type":49,"tag":86,"props":672,"children":674},{"emptyLinePlaceholder":673},true,[675],{"type":55,"value":676},"\n",{"type":49,"tag":86,"props":678,"children":680},{"class":88,"line":679},6,[681],{"type":49,"tag":86,"props":682,"children":683},{},[684],{"type":55,"value":685},"gap = df[(df['HbA1c'] > 9) & (~df['on_target_med'])]\n",{"type":49,"tag":86,"props":687,"children":689},{"class":88,"line":688},7,[690],{"type":49,"tag":86,"props":691,"children":692},{},[693],{"type":55,"value":694},"denom = len(df[df['HbA1c'].notna()])\n",{"type":49,"tag":86,"props":696,"children":698},{"class":88,"line":697},8,[699],{"type":49,"tag":86,"props":700,"children":701},{},[702],{"type":55,"value":703},"pct = f\"{len(gap)\u002Fdenom*100:.1f}%\" if denom > 0 else \"N\u002FA (no HbA1c data)\"\n",{"type":49,"tag":86,"props":705,"children":707},{"class":88,"line":706},9,[708],{"type":49,"tag":86,"props":709,"children":710},{},[711],{"type":55,"value":712},"print(f\"Care gap: {len(gap)}\u002F{denom} ({pct})\")\n",{"type":49,"tag":58,"props":714,"children":716},{"id":715},"visualization",[717],{"type":55,"value":718},"Visualization",{"type":49,"tag":65,"props":720,"children":721},{},[722,724,730,732,738],{"type":55,"value":723},"Always use dark theme. Use ",{"type":49,"tag":82,"props":725,"children":727},{"className":726},[],[728],{"type":55,"value":729},"save_chart_to_canvas()",{"type":55,"value":731}," instead of ",{"type":49,"tag":82,"props":733,"children":735},{"className":734},[],[736],{"type":55,"value":737},"plt.savefig()",{"type":55,"value":739}," directly.",{"type":49,"tag":75,"props":741,"children":743},{"className":77,"code":742,"language":22,"meta":79,"style":79},"import matplotlib\nmatplotlib.use('Agg')\nimport matplotlib.pyplot as plt\n\nplt.style.use('dark_background')\nfig, ax = plt.subplots(figsize=(10, 6))\nfig.patch.set_facecolor('#1a1a1a')\nax.set_facecolor('#1a1a1a')\n\n# Histogram with NVIDIA green\nax.hist(values, bins=15, color='#76B900', edgecolor='#1a1a1a', alpha=0.85)\nax.axvline(x=threshold, color='#ff4444', linestyle='--', linewidth=2, label=f'Threshold ({threshold})')\nax.set_title(\"Title\", fontsize=14, fontweight='bold', color='white')\nax.legend()\nax.grid(axis='y', alpha=0.2, color='#444444')\nax.text(0.98, 0.95, f\"N = {len(values)}\", transform=ax.transAxes, fontsize=11, color='#888888', ha='right', va='top')\n\n# MANDATORY: use save_chart_to_canvas (NOT plt.savefig)\nsave_chart_to_canvas(fig, \"chart.png\")\nplt.close()\n",[744],{"type":49,"tag":82,"props":745,"children":746},{"__ignoreMap":79},[747,755,763,771,778,786,794,802,810,817,826,835,844,853,862,871,880,888,897,906],{"type":49,"tag":86,"props":748,"children":749},{"class":88,"line":89},[750],{"type":49,"tag":86,"props":751,"children":752},{},[753],{"type":55,"value":754},"import matplotlib\n",{"type":49,"tag":86,"props":756,"children":757},{"class":88,"line":98},[758],{"type":49,"tag":86,"props":759,"children":760},{},[761],{"type":55,"value":762},"matplotlib.use('Agg')\n",{"type":49,"tag":86,"props":764,"children":765},{"class":88,"line":107},[766],{"type":49,"tag":86,"props":767,"children":768},{},[769],{"type":55,"value":770},"import matplotlib.pyplot as plt\n",{"type":49,"tag":86,"props":772,"children":773},{"class":88,"line":660},[774],{"type":49,"tag":86,"props":775,"children":776},{"emptyLinePlaceholder":673},[777],{"type":55,"value":676},{"type":49,"tag":86,"props":779,"children":780},{"class":88,"line":669},[781],{"type":49,"tag":86,"props":782,"children":783},{},[784],{"type":55,"value":785},"plt.style.use('dark_background')\n",{"type":49,"tag":86,"props":787,"children":788},{"class":88,"line":679},[789],{"type":49,"tag":86,"props":790,"children":791},{},[792],{"type":55,"value":793},"fig, ax = plt.subplots(figsize=(10, 6))\n",{"type":49,"tag":86,"props":795,"children":796},{"class":88,"line":688},[797],{"type":49,"tag":86,"props":798,"children":799},{},[800],{"type":55,"value":801},"fig.patch.set_facecolor('#1a1a1a')\n",{"type":49,"tag":86,"props":803,"children":804},{"class":88,"line":697},[805],{"type":49,"tag":86,"props":806,"children":807},{},[808],{"type":55,"value":809},"ax.set_facecolor('#1a1a1a')\n",{"type":49,"tag":86,"props":811,"children":812},{"class":88,"line":706},[813],{"type":49,"tag":86,"props":814,"children":815},{"emptyLinePlaceholder":673},[816],{"type":55,"value":676},{"type":49,"tag":86,"props":818,"children":820},{"class":88,"line":819},10,[821],{"type":49,"tag":86,"props":822,"children":823},{},[824],{"type":55,"value":825},"# Histogram with NVIDIA green\n",{"type":49,"tag":86,"props":827,"children":829},{"class":88,"line":828},11,[830],{"type":49,"tag":86,"props":831,"children":832},{},[833],{"type":55,"value":834},"ax.hist(values, bins=15, color='#76B900', edgecolor='#1a1a1a', alpha=0.85)\n",{"type":49,"tag":86,"props":836,"children":838},{"class":88,"line":837},12,[839],{"type":49,"tag":86,"props":840,"children":841},{},[842],{"type":55,"value":843},"ax.axvline(x=threshold, color='#ff4444', linestyle='--', linewidth=2, label=f'Threshold ({threshold})')\n",{"type":49,"tag":86,"props":845,"children":847},{"class":88,"line":846},13,[848],{"type":49,"tag":86,"props":849,"children":850},{},[851],{"type":55,"value":852},"ax.set_title(\"Title\", fontsize=14, fontweight='bold', color='white')\n",{"type":49,"tag":86,"props":854,"children":856},{"class":88,"line":855},14,[857],{"type":49,"tag":86,"props":858,"children":859},{},[860],{"type":55,"value":861},"ax.legend()\n",{"type":49,"tag":86,"props":863,"children":865},{"class":88,"line":864},15,[866],{"type":49,"tag":86,"props":867,"children":868},{},[869],{"type":55,"value":870},"ax.grid(axis='y', alpha=0.2, color='#444444')\n",{"type":49,"tag":86,"props":872,"children":874},{"class":88,"line":873},16,[875],{"type":49,"tag":86,"props":876,"children":877},{},[878],{"type":55,"value":879},"ax.text(0.98, 0.95, f\"N = {len(values)}\", transform=ax.transAxes, fontsize=11, color='#888888', ha='right', va='top')\n",{"type":49,"tag":86,"props":881,"children":883},{"class":88,"line":882},17,[884],{"type":49,"tag":86,"props":885,"children":886},{"emptyLinePlaceholder":673},[887],{"type":55,"value":676},{"type":49,"tag":86,"props":889,"children":891},{"class":88,"line":890},18,[892],{"type":49,"tag":86,"props":893,"children":894},{},[895],{"type":55,"value":896},"# MANDATORY: use save_chart_to_canvas (NOT plt.savefig)\n",{"type":49,"tag":86,"props":898,"children":900},{"class":88,"line":899},19,[901],{"type":49,"tag":86,"props":902,"children":903},{},[904],{"type":55,"value":905},"save_chart_to_canvas(fig, \"chart.png\")\n",{"type":49,"tag":86,"props":907,"children":909},{"class":88,"line":908},20,[910],{"type":49,"tag":86,"props":911,"children":912},{},[913],{"type":55,"value":914},"plt.close()\n",{"type":49,"tag":58,"props":916,"children":918},{"id":917},"guardrails",[919],{"type":55,"value":920},"Guardrails",{"type":49,"tag":407,"props":922,"children":923},{},[924,929,934,939,944],{"type":49,"tag":411,"props":925,"children":926},{},[927],{"type":55,"value":928},"Never compute statistics on fewer than 5 data points",{"type":49,"tag":411,"props":930,"children":931},{},[932],{"type":55,"value":933},"Always report sample size: \"45.0% (27 out of 60)\"",{"type":49,"tag":411,"props":935,"children":936},{},[937],{"type":55,"value":938},"Flag data quality issues if >30% missing",{"type":49,"tag":411,"props":940,"children":941},{},[942],{"type":55,"value":943},"Do not fabricate data — report what exists, flag what's missing",{"type":49,"tag":411,"props":945,"children":946},{},[947],{"type":55,"value":948},"All charts must include N annotation",{"type":49,"tag":58,"props":950,"children":952},{"id":951},"output-format",[953],{"type":55,"value":954},"Output Format",{"type":49,"tag":65,"props":956,"children":957},{},[958],{"type":55,"value":959},"End every script with:",{"type":49,"tag":75,"props":961,"children":963},{"className":77,"code":962,"language":22,"meta":79,"style":79},"print(f\"\\nDisclaimer: This analysis is for research and operational purposes.\")\nprint(\"Clinical decisions should be made by qualified clinicians.\")\n",[964],{"type":49,"tag":82,"props":965,"children":966},{"__ignoreMap":79},[967,975],{"type":49,"tag":86,"props":968,"children":969},{"class":88,"line":89},[970],{"type":49,"tag":86,"props":971,"children":972},{},[973],{"type":55,"value":974},"print(f\"\\nDisclaimer: This analysis is for research and operational purposes.\")\n",{"type":49,"tag":86,"props":976,"children":977},{"class":88,"line":98},[978],{"type":49,"tag":86,"props":979,"children":980},{},[981],{"type":55,"value":982},"print(\"Clinical decisions should be made by qualified clinicians.\")\n",{"type":49,"tag":984,"props":985,"children":986},"style",{},[987],{"type":55,"value":988},"html .light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html.light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"items":990,"total":1147},[991,1009,1025,1036,1048,1062,1075,1089,1102,1113,1127,1136],{"slug":992,"name":992,"fn":993,"description":994,"org":995,"tags":996,"stars":1006,"repoUrl":1007,"updatedAt":1008},"nemoclaw-user-guide","retrieve NemoClaw documentation and configuration","Guides human users' AI agents to the NemoClaw docs MCP server and canonical Fern documentation in Markdown form. Use when users ask how to install, configure, operate, troubleshoot, secure, or learn NemoClaw with an AI coding assistant. Trigger keywords - nemoclaw docs, use nemoclaw with ai agent, nemoclaw mcp docs, nemoclaw install help, nemoclaw quickstart, nemoclaw markdown docs, llms.txt, agent skills.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[997,1000,1003],{"name":998,"slug":999,"type":15},"Documentation","documentation",{"name":1001,"slug":1002,"type":15},"MCP","mcp",{"name":1004,"slug":1005,"type":15},"Search","search",21777,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNemoClaw","2026-07-20T06:00:01.461044",{"slug":1010,"name":1010,"fn":1011,"description":1012,"org":1013,"tags":1014,"stars":1022,"repoUrl":1023,"updatedAt":1024},"mcore-build-and-dependency","manage Megatron-LM development environments","Container-based dev environment setup and dependency management for Megatron-LM. Covers acquiring and launching the CI container, uv package management, and updating uv.lock.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1015,1018,1021],{"name":1016,"slug":1017,"type":15},"Containers","containers",{"name":1019,"slug":1020,"type":15},"Deployment","deployment",{"name":21,"slug":22,"type":15},17049,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM","2026-07-27T06:06:11.249662",{"slug":1026,"name":1026,"fn":1027,"description":1028,"org":1029,"tags":1030,"stars":1022,"repoUrl":1023,"updatedAt":1035},"mcore-bump-base-image","update NVIDIA PyTorch base images","Bump the NVIDIA PyTorch base image (`nvcr.io\u002Fnvidia\u002Fpytorch:YY.MM-py3`) used by Megatron-LM CI. Covers the two pin sites (GitHub CI in `docker\u002F.ngc_version.dev` and GitLab CI in `.gitlab\u002Fstages\u002F01.build.yml`), the post-bump CI loop (re-run functional tests, refresh golden values, mark broken tests), and the gotchas that bit PRs",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1031,1034],{"name":1032,"slug":1033,"type":15},"CI\u002FCD","ci-cd",{"name":1019,"slug":1020,"type":15},"2026-07-14T05:25:59.97109",{"slug":1037,"name":1037,"fn":1038,"description":1039,"org":1040,"tags":1041,"stars":1022,"repoUrl":1023,"updatedAt":1047},"mcore-cicd","manage CI\u002FCD pipelines for Megatron-LM","CI\u002FCD reference for Megatron-LM. Covers CI pipeline structure, PR scope labels, triggering internal GitLab CI (which force-pushes the current branch to a pull-request\u002FBRANCH ref — always dry-run and verify the destination first; never run against shared or protected branches), and CI failure investigation.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1042,1043,1044],{"name":1032,"slug":1033,"type":15},{"name":1019,"slug":1020,"type":15},{"name":1045,"slug":1046,"type":15},"GitHub","github","2026-07-27T06:06:12.278222",{"slug":1049,"name":1049,"fn":1050,"description":1051,"org":1052,"tags":1053,"stars":1022,"repoUrl":1023,"updatedAt":1061},"mcore-create-issue","investigate CI failures and create issues","Investigate a failing GitHub Actions run or job and create a GitHub issue for the failure.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1054,1057,1058],{"name":1055,"slug":1056,"type":15},"Debugging","debugging",{"name":1045,"slug":1046,"type":15},{"name":1059,"slug":1060,"type":15},"Triage","triage","2026-07-14T05:25:57.442089",{"slug":1063,"name":1063,"fn":1064,"description":1065,"org":1066,"tags":1067,"stars":1022,"repoUrl":1023,"updatedAt":1074},"mcore-linting-and-formatting","lint and format Megatron-LM code","Linting and formatting for Megatron-LM. Covers running autoformat.sh, tools (ruff, black, isort, pylint, mypy), and code style rules.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1068,1071],{"name":1069,"slug":1070,"type":15},"Best Practices","best-practices",{"name":1072,"slug":1073,"type":15},"Code Analysis","code-analysis","2026-07-14T05:25:56.18433",{"slug":1076,"name":1076,"fn":1077,"description":1078,"org":1079,"tags":1080,"stars":1022,"repoUrl":1023,"updatedAt":1088},"mcore-migrate-gpt-to-hybrid","migrate Megatron-LM models to HybridModel","Migration guide for moving Megatron Core GPTModel checkpoints, model providers, training commands, and layer mappings to HybridModel.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1081,1084,1087],{"name":1082,"slug":1083,"type":15},"Machine Learning","machine-learning",{"name":1085,"slug":1086,"type":15},"Migration","migration",{"name":9,"slug":8,"type":15},"2026-07-17T06:07:11.777011",{"slug":1090,"name":1090,"fn":1091,"description":1092,"org":1093,"tags":1094,"stars":1022,"repoUrl":1023,"updatedAt":1101},"mcore-onboard-gb200-1node-tests","onboard functional tests for GB200","Onboard 1-node GitHub MR functional tests for GB200 from existing mr-scoped 2-node tests.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1095,1098],{"name":1096,"slug":1097,"type":15},"QA","qa",{"name":1099,"slug":1100,"type":15},"Testing","testing","2026-07-14T05:25:53.673039",{"slug":1103,"name":1103,"fn":1104,"description":1105,"org":1106,"tags":1107,"stars":1022,"repoUrl":1023,"updatedAt":1112},"mcore-run-on-slurm","launch distributed training jobs on SLURM","How to launch distributed Megatron-LM training jobs on a SLURM cluster. Covers a minimal sbatch skeleton, environment-variable setup for torch.distributed.run, CUDA_DEVICE_MAX_CONNECTIONS rules across hardware and parallelism modes, container conventions, monitoring, and per-rank failure diagnosis.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1108,1109],{"name":1019,"slug":1020,"type":15},{"name":1110,"slug":1111,"type":15},"Infrastructure","infrastructure","2026-07-14T05:25:49.362534",{"slug":1114,"name":1114,"fn":1115,"description":1116,"org":1117,"tags":1118,"stars":1022,"repoUrl":1023,"updatedAt":1126},"mcore-split-pr","split pull requests to reduce review load","Split a PR into multiple PRs to reduce the number of required CODEOWNERS reviewer groups.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1119,1122,1123],{"name":1120,"slug":1121,"type":15},"Code Review","code-review",{"name":1045,"slug":1046,"type":15},{"name":1124,"slug":1125,"type":15},"Pull Requests","pull-requests","2026-07-14T05:26:01.226578",{"slug":1128,"name":1128,"fn":1129,"description":1130,"org":1131,"tags":1132,"stars":1022,"repoUrl":1023,"updatedAt":1135},"mcore-testing","run and manage Megatron-LM tests","Test system for Megatron-LM. Covers test layout, recipe YAML structure, adding and running unit and functional tests, golden values, marker filters, and CI parity.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1133,1134],{"name":1096,"slug":1097,"type":15},{"name":1099,"slug":1100,"type":15},"2026-07-14T05:25:54.928983",{"slug":1137,"name":1137,"fn":1138,"description":1139,"org":1140,"tags":1141,"stars":1022,"repoUrl":1023,"updatedAt":1146},"nightly-sync","manage nightly main-to-dev sync workflows","Domain knowledge for the nightly main-to-dev sync workflow. Covers merge strategy, CI architecture, failure investigation, and known issues.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1142,1145],{"name":1143,"slug":1144,"type":15},"Automation","automation",{"name":1032,"slug":1033,"type":15},"2026-07-30T05:29:03.275638",496,{"items":1149,"total":828},[1150,1158,1171,1186,1201,1212,1227],{"slug":4,"name":4,"fn":5,"description":6,"org":1151,"tags":1152,"stars":26,"repoUrl":27,"updatedAt":28},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1153,1154,1155,1156,1157],{"name":17,"slug":18,"type":15},{"name":24,"slug":25,"type":15},{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},{"name":21,"slug":22,"type":15},{"slug":1159,"name":1159,"fn":1160,"description":1161,"org":1162,"tags":1163,"stars":26,"repoUrl":27,"updatedAt":1170},"case-summary","summarize clinical patient cases from FHIR","Prepare a complete clinical case summary for a patient from FHIR endpoints. Use when asked to summarize a patient, compile a case, or prepare for tumor board.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1164,1165,1166,1167],{"name":24,"slug":25,"type":15},{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},{"name":1168,"slug":1169,"type":15},"Summarization","summarization","2026-07-14T05:35:52.790528",{"slug":1172,"name":1172,"fn":1173,"description":1174,"org":1175,"tags":1176,"stars":26,"repoUrl":27,"updatedAt":1185},"clinical-delegation","delegate clinical tasks to specialist agents","How to delegate clinical tasks to specialist agents. Always use sub-agent runtime with explicit agentId — never ACP. Never call FHIR via web_fetch.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1177,1180,1181,1184],{"name":1178,"slug":1179,"type":15},"Agents","agents",{"name":13,"slug":14,"type":15},{"name":1182,"slug":1183,"type":15},"Multi-Agent","multi-agent",{"name":9,"slug":8,"type":15},"2026-07-14T05:35:55.294972",{"slug":1187,"name":1187,"fn":1188,"description":1189,"org":1190,"tags":1191,"stars":26,"repoUrl":27,"updatedAt":1200},"clinical-knowledge","provide clinical reference and regulatory context","Teaches agents clinical reference ranges, condition codes, quality measure definitions, drug classifications, and regulatory context so they can flag abnormal values and identify care gaps.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1192,1195,1196,1197],{"name":1193,"slug":1194,"type":15},"Clinical Trials","clinical-trials",{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},{"name":1198,"slug":1199,"type":15},"Regulatory Compliance","regulatory-compliance","2026-07-14T05:35:56.550833",{"slug":1202,"name":1202,"fn":1203,"description":1204,"org":1205,"tags":1206,"stars":26,"repoUrl":27,"updatedAt":1211},"cohort-compare","analyze patient cohorts from FHIR endpoints","Analyze a cohort of patients from FHIR endpoints to find care gaps and patterns. Use when asked to compare patients, find quality gaps, or analyze a population.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1207,1208,1209,1210],{"name":17,"slug":18,"type":15},{"name":24,"slug":25,"type":15},{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},"2026-07-14T05:35:51.543095",{"slug":1213,"name":1213,"fn":1214,"description":1215,"org":1216,"tags":1217,"stars":26,"repoUrl":27,"updatedAt":1226},"dgx-diagnose","diagnose NVIDIA DGX Station hardware issues","Diagnose common DGX Station GB300 issues — CUDA crashes, wrong-GPU targeting, vLLM\u002FSGLang container bugs, MIG state problems, NVLink\u002FFabric Manager errors, X\u002FVulkan failures, HuggingFace auth, and port conflicts. Use when the user reports a GPU error, inference server crash, MIG problem, or any unexplained DGX Station failure.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1218,1221,1222,1223],{"name":1219,"slug":1220,"type":15},"AI Infrastructure","ai-infrastructure",{"name":1055,"slug":1056,"type":15},{"name":9,"slug":8,"type":15},{"name":1224,"slug":1225,"type":15},"Observability","observability","2026-07-14T05:31:04.085598",{"slug":1228,"name":1228,"fn":1229,"description":1230,"org":1231,"tags":1232,"stars":26,"repoUrl":27,"updatedAt":1239},"fhir-basics","query and parse FHIR R4 API resources","Teaches agents how FHIR R4 APIs work, what resources are available, how to query them with search parameters, and how to correctly parse all response formats including component Observations.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1233,1236,1237,1238],{"name":1234,"slug":1235,"type":15},"API Development","api-development",{"name":24,"slug":25,"type":15},{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},"2026-07-14T05:35:57.797226"]