Mapping To Snomed

What it does

When your agent uses it

• The user wants to code findings
• Body structures
• Substances to SNOMED CT
• Run an ECL query

Example prompts

• “/mapping-to-snomed”

Requirements

• Python 3
• A credential in FHIR_TX_TOKEN

Workflow steps

7 steps, taken from the first numbered list in SKILL.md.

1. 1Extract spans with OpenMed (Disease, Anatomy, Pharmaceutical models).
2. 2Pick a semantic constraint (ECL) from the OpenMed label so you search the
3. 3Search with ValueSet/$expand?filter= under that ECL.
4. 4Rank & disambiguate by display match and confidence; prefer the most
5. 5Validate with $validate-code; $lookup to capture the FSN and any
6. 6Translate instead of searching when you already hold an ICD-10/local code
7. 7Emit {system: "http://snomed.info/sct", code, display} — the SCTID plus

What it can do on your machine

Read from SKILL.md and the folder at commit 34d7b8c. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  No scripts in the folder and no shell commands in SKILL.md (its code samples are python).

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  Hosts in commands or code, which the agent is likely to contact:

  • snomed.info

  Also links to:

  • snomed.org
  • nlm.nih.gov
  • confluence.ihtsdotools.org
  • hl7.org
  • github.com
  • ontoserver.csiro.au

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names these keys or tokens, usually read from environment variables:

  • FHIR_TX_TOKEN

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Mapping To Snomed loads about 2.1k tokens when it runs. Until then it costs about 211 tokens; SKILL.md has 650 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~211

When it runs · the whole SKILL.md, loaded when a task matches

~2.1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from maziyarpanahi/openmed at commit 34d7b8c, republished under its Apache-2.0 licence (© maziyarpanahi). 650 words, ~2,147 tokens.

name

mapping-to-snomed

description

Maps clinical concept spans extracted by OpenMed to SNOMED CT concepts through a USER-SUPPLIED terminology server (the user's own Ontoserver, Snowstorm, or UMLS/UTS), never a bundled vocabulary. Use when the user wants to code findings, disorders, procedures, body structures, or substances to SNOMED CT, run an ECL query, translate via a ConceptMap, or resolve a span to a concept id with FHIR $lookup/$translate/$validate-code. Trigger keywords: SNOMED CT, SNOMED concept id, ECL, ConceptMap, $translate, $lookup, Ontoserver, Snowstorm, SCTID, post-coordination, terminology server. Pairs after OpenMed NER: consume Disease/Anatomy/Pharmaceutical entities from openmed.analyze_text and map each span out-of-process. SNOMED CT is license-restricted — it is NEVER bundled; the user calls their own affiliate-licensed server.

license

Apache-2.0

metadata.project

OpenMed

metadata.category

terminology-coding

metadata.pairs

after

metadata.version

1.0

import os, requests

# Provided by the USER — their licensed server. Nothing bundled.
TX = os.environ["FHIR_TX_URL"]              # e.g. https://snowstorm.example.org/fhir
TOKEN = os.environ.get("FHIR_TX_TOKEN")     # if the server requires auth
SNOMED = "http://snomed.info/sct"
HDRS = {"Accept": "application/fhir+json"}
if TOKEN:
    HDRS["Authorization"] = f"Bearer {TOKEN}"

def lookup(code: str) -> dict:
    """$lookup: fully specified name + properties for an SCTID."""
    r = requests.get(f"{TX}/CodeSystem/$lookup",
                     params={"system": SNOMED, "code": code},
                     headers=HDRS, timeout=15)
    r.raise_for_status()
    return r.json()

def find_concepts(text: str, ecl: str = "<<404684003", count: int = 10):
    """Text search constrained by ECL (default: descendants of Clinical finding)."""
    vs = f"{SNOMED}?fhir_vs=ecl/{ecl}"
    r = requests.get(f"{TX}/ValueSet/$expand",
                     params={"url": vs, "filter": text, "count": count},
                     headers=HDRS, timeout=20)
    r.raise_for_status()
    return r.json().get("expansion", {}).get("contains", [])

def translate(code: str, source_system: str, conceptmap_url: str):
    """$translate an existing code to SNOMED CT via a ConceptMap."""
    r = requests.get(f"{TX}/ConceptMap/$translate",
                     params={"url": conceptmap_url, "system": source_system,
                             "code": code, "targetsystem": SNOMED},
                     headers=HDRS, timeout=20)
    r.raise_for_status()
    return r.json()

# ECL examples: 64572001=disease, 71388002=procedure, 123037004=body structure
print(find_concepts("type 2 diabetes", ecl="<<64572001"))

import openmed

note = "Assessment: type 2 diabetes mellitus with diabetic nephropathy."
result = openmed.analyze_text(
    note,
    model_name="disease_detection_superclinical",   # Disease category
    output_format="dict",
)

ECL_FOR_LABEL = {
    "DISEASE":   "<<64572001",     # | Disease |
    "CONDITION": "<<64572001",
    "PATHOLOGY": "<<64572001",
    "ANATOMY":   "<<123037004",    # | Body structure |
    "ORGAN":     "<<123037004",
}

for ent in result["entities"]:
    ecl = ECL_FOR_LABEL.get(ent["label"], "<<404684003")  # fallback: Clinical finding
    candidates = find_concepts(ent["text"], ecl=ecl, count=5)
    print(ent["text"], ent["start"], ent["end"], "->",
          [(c["code"], c["display"]) for c in candidates[:3]])

© maziyarpanahi, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/mapping-to-snomed of maziyarpanahi/openmed.

Open the folder on GitHubat commit 34d7b8c