Reconciling Problem Lists

What it does

When your agent uses it

• Wants a problem list
• Condition reconciliation
• Dedup of synonymous diagnosis mentions
• Active-vs-resolved status from a note

Example prompts

• “/reconciling-problem-lists”

Requirements

• Python 3

Workflow steps

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

1. 1Collect Disease/Condition entities from analyze_text across the whole
2. 2Attach clinical context per mention with resolve_span_context (or the
3. 3Exclude what isn't a problem. Drop NEGATED mentions (patient denies /
4. 4Cluster synonymous mentions into one concept. Fold surface variants
5. 5Assign status by aggregating context. A concept that is RECENT/active
6. 6Emit the reconciled list — one record per concept with status, mention

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):

  • healthit.gov
  • hl7.org
  • snomed.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

Reconciling Problem Lists loads about 1.8k tokens when it runs. Until then it costs about 189 tokens; SKILL.md has 570 words of instructions outside code blocks.

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

~189

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). 570 words, ~1,770 tokens.

name

reconciling-problem-lists

description

Deduplicate and reconcile OpenMed-extracted conditions into one clean active problem list with clinical status (active / resolved / historical). Use after NER and context resolution when the user wants a problem list, condition reconciliation, dedup of synonymous diagnosis mentions, or active-vs-resolved status from a note. Covers clustering synonymous mentions into one concept, excluding negated mentions, applying clinical context (historical / hypothetical / recent) to set status, and emitting a USCDI-Problem-shaped list. SNOMED CT concept grounding is user-supplied and out-of-process. Hand-off: consume openmed.analyze_text Disease entities plus resolving-clinical-context axes. 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, NEGATED, HISTORICAL, HYPOTHETICAL

note = ("PMH: type 2 diabetes, prior MI 2019 (resolved). "
        "A&P: poorly controlled DM2; denies chest pain.")

ents = openmed.analyze_text(note, model_name="disease_detection_superclinical",
                            output_format="dict")

def normalize(surface: str) -> str:
    # Cheap synonym folding; replace with SNOMED grounding (out-of-process).
    s = surface.lower().strip()
    return {"dm2": "type 2 diabetes", "diabetes mellitus": "type 2 diabetes"}.get(s, s)

problems = {}  # concept -> reconciled record
for e in ents:
    surface = e["word"]
    ctx = resolve_span_context(surface, note)
    if ctx.negation == NEGATED:
        continue                                   # patient does NOT have it -> exclude
    concept = normalize(surface)
    status = ("resolved" if ctx.temporality == HISTORICAL else
              "active")
    if ctx.temporality == HYPOTHETICAL:
        continue                                   # not asserted as present
    rec = problems.setdefault(concept, {"concept": concept, "status": status,
                                        "mentions": 0})
    rec["mentions"] += 1
    # Active anywhere wins over a historical mention of the same concept.
    if status == "active":
        rec["status"] = "active"

problem_list = list(problems.values())
# -> [{"concept": "type 2 diabetes", "status": "active", "mentions": 2}, ...]
# "chest pain" excluded (negated); "MI" -> historical/resolved.

© 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/reconciling-problem-lists of maziyarpanahi/openmed.

Open the folder on GitHubat commit 9dca507