Treatment-Plan Documentation Validator
A documentation-only skill that formats and structurally validates treatment-plan records after licensed clinicians have already made and verified every decision.
Docs & OfficeAdvanced★ 33,030⑂ 3,248AI score 7/10Last updated: Aug 9, 2026
What it does
This skill takes clinical decisions that have already been made and verified by authorized licensed professionals and helps you transcribe them into six standard JSON templates, then validates the package structurally using local, offline Python scripts.
- Generates templates for source-fact manifests, clinician-authored interventions, goals/monitoring checkpoints, shared-decision records, transition reconciliation, and intended-use handoff
- Runs deterministic checks for traceability, completeness, consistency, and privacy-process handling
- Builds a schedule only from dates already present in the package — it never infers intervals or recurrence
- Keeps a mandatory "DRAFT — NOT MEDICAL ADVICE — CLINICIAN SIGN-OFF REQUIRED" notice and holds the release gate
blockeduntil human review completes
It explicitly refuses diagnosis, therapy selection, dosing, interaction checking, triage, and prognosis.
Who it's for
- Clinical documentation and quality-assurance teams in hospitals or clinics
- Developers building health-record or EMR documentation pipelines
- Teams that want to catch missing provenance in handoff and medication-reconciliation records
- Research or training groups designing documentation workflows with synthetic data
Examples
- Bootstrap a synthetic package: run
generate_template.py --subject-ref SYNTHETIC-CASE-001 --classification syntheticto produce all six templates for training material. - Audit a handoff record: run
check_completeness.pyandvalidate_traceability.pyto list field paths missing source locators, verifier roles, or verification timestamps. - Produce an explicit date schedule: feed clinician-supplied checkpoint dates into
timeline_generator.pyto emitexplicit-date-schedule.jsonwhile the draft notice remains attached until sign-off.
· · · Install guide · · ·
Install in the Claude app (no terminal)
- Download the ZIP with the button below.
- In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
- Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
Install in Claude Code
Let Claude do it — paste this into Claude Code
Install the skill I found on Claude Skill Mart. Copy the skills/treatment-plans folder from the GitHub repo K-Dense-AI/scientific-agent-skills into my ~/.claude/skills/treatment-plans/. When it's done, tell me in one line what this skill can do.
Install with a command instead
git clone https://github.com/K-Dense-AI/scientific-agent-skills.git /tmp/scientific-agent-skills && mkdir -p ~/.claude/skills && cp -r /tmp/scientific-agent-skills/skills/treatment-plans ~/.claude/skills/⚠ This is a third-party skill. Check the source repository before installing.
- Open a terminal (Terminal on macOS/Linux, WSL or Git Bash on Windows).
- Confirm Python 3.11 or newer:
python3 --version - Clone the repository:
git clone https://github.com/K-Dense-AI/scientific-agent-skills.git - Create the skills directory:
mkdir -p ~/.claude/skills - Copy just this skill:
cp -r scientific-agent-skills/skills/treatment-plans ~/.claude/skills/ - Restart Claude Code and try a prompt such as "validate my treatment-plan documentation package".
- (Optional) Run the bundled tests:
PYTHONDONTWRITEBYTECODE=1 python3 -m unittest discover -s tests/treatment-plans -p 'test_*.py' -v - For any real patient-derived data, work only inside an institutionally authorized environment and use the minimum necessary information.
View source on GitHub ↗License: MIT