Resolving Clinical Context

What it does

When your agent uses it

• Needs assertion status
• Negation detection
• Family-history / hypothetical / historical flags
• ConText/NegEx-style classification before grounding entities to FHIR

Example prompts

• “denies chest pain”
• “history of MI”
• “/resolving-clinical-context”

Requirements

• Python 3

Workflow steps

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

1. 1Get entities and their context window. From analyze_text, take each
2. 2Resolve negation with resolve_negation(span, window) → AFFIRMED or
3. 3Resolve temporality with resolve_temporality(span, window) → RECENT
4. 4Resolve uncertainty with resolve_uncertainty(span, window) → CERTAIN
5. 5Apply the axes downstream. Drop or refute NEGATED spans; route

What it can do on your machine

Read from SKILL.md and the folder at commit 9dca507. 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

  Links to these hosts (documentation or services it may open):

  • doi.org
  • hl7.org

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

• Credentials

  Names no API keys, tokens, secrets or passwords.

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

Context cost

Resolving Clinical Context loads about 1.8k tokens when it runs. Until then it costs about 183 tokens; SKILL.md has 606 words of instructions outside code blocks.

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

~183

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

~1.8k

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 9dca507, republished under its Apache-2.0 licence (© maziyarpanahi). 606 words, ~1,762 tokens.

name

resolving-clinical-context

description

Assign negation, temporality, and uncertainty (the ConText axes) to clinical entities extracted by OpenMed, so "denies chest pain" is not counted as chest pain and "history of MI" is not counted as an active MI. Use after NER when the user needs assertion status, negation detection, family-history / hypothetical / historical flags, or ConText/NegEx-style classification before grounding entities to FHIR or a problem list. Covers openmed.clinical.resolve_negation / resolve_temporality / resolve_uncertainty / resolve_span_context / assert_context_axes, ClinicalAssertion, and the AFFIRMED/NEGATED, RECENT/HISTORICAL/HYPOTHETICAL, CERTAIN/UNCERTAIN constants. Pairs after extracting-clinical-entities.

license

Apache-2.0

metadata.project

OpenMed

metadata.category

clinical-nlp

metadata.pairs

after

metadata.version

1.0

import openmed
from openmed.clinical import (
    resolve_span_context, assert_context_axes,
    NEGATED, HISTORICAL, HYPOTHETICAL, UNCERTAIN,
)

note = "Patient denies chest pain. History of MI. Concern for PE; rule out DVT."

# 1) Extract entities (registry key, HF id, or local path).
ents = openmed.analyze_text(note, model_name="disease_detection_superclinical",
                            output_format="dict")

# 2) Assign ConText axes per entity. Pass the span text plus a window of cues.
for e in ents:
    span = e["word"]                      # entity surface text
    window = note                         # full sentence/note as modifier context
    ctx = resolve_span_context(span, window)
    print(span, "->", ctx.negation, ctx.temporality, ctx.certainty)

# "chest pain" -> negated   recent      certain     (do NOT record as present)
# "MI"         -> affirmed  historical  certain     (past, not active)
# "PE"         -> affirmed  recent      uncertain   (hedged; flag, don't drop)

© 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/resolving-clinical-context of maziyarpanahi/openmed.

Open the folder on GitHubat commit 9dca507