OpenMed is local-first healthcare AI software that extracts clinical information and removes personally identifying details from clinical text on hardware controlled by the user. Healthcare developers use its Python runtime, Apple Silicon and mobile SDKs, and browser support for on-device clinical NER and PII de-identification.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add maziyarpanahi/openmed --skill configuring-privacy-policiesgit clone --depth 1 https://github.com/maziyarpanahi/openmedWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/maziyarpanahi/openmed/configuring-privacy-policies)<a href="https://agentmods.dev/skills/maziyarpanahi/openmed/configuring-privacy-policies"><img src="https://agentmods.dev/badge/skills/maziyarpanahi/openmed/configuring-privacy-policies/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/maziyarpanahi/openmed/configuring-privacy-policies"><img src="https://agentmods.dev/badge/skills/maziyarpanahi/openmed/configuring-privacy-policies.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00143 | $0.02155 |
| Opus 5 | $0.00072 | $0.01077 |
| Sonnet 5 | $0.00029 | $0.00431 |
| Haiku 4.5 | $0.00014 | $0.00215 |
Grade A, and why
configuring-privacy-policies scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 156 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Configuring privacy policies
A policy profile is a named bundle of de-identification decisions: which
action (mask/redact/replace/keep) applies to each label, how aggressively
detectors arbitrate, whether the mandatory safety sweep runs, and whether a
reversible mapping is produced. OpenMed ships seven profiles. Pass one by name
to deidentify(policy=...) and you get a compliance-aligned default without
hand-wiring 50+ per-label actions. Everything runs on-device.
When to use this skill
Use it to pick the right policy= for a regulatory context, to understand what
a profile actually changes, or to go beyond the bundle — keeping quasi-
identifiers for research, or registering a custom surrogate generator (e.g. your
own MRN format).
Quick start
import openmed
note = "Jane Roe, DOB 1979-04-11, lives in Cambridge MA 02139. SSN 123-45-6789."
# HIPAA Safe Harbor: mask every identifier class.
safe = openmed.deidentify(note, policy="hipaa_safe_harbor")
# GDPR pseudonymization: replace with fakes AND keep a reversible mapping.
gdpr = openmed.deidentify(note, policy="gdpr_pseudonymization")
mapping = gdpr.mapping # present because the profile sets keep_mapping=True
# Research limited dataset: mask direct identifiers, KEEP quasi-identifiers
# (dates, age, ZIP, geography) so the data stays analytically useful.
lds = openmed.deidentify(note, policy="research_limited_dataset")
The seven bundled profiles
Each profile lives in openmed/core/policies/<name>.json. Summary of what each
actually configures:
| Profile | Default action | Quasi-identifiers | Mapping | Safety sweep | Use case |
|---|---|---|---|---|---|
hipaa_safe_harbor |
mask all | masked | none | mandatory | HIPAA §164.514(b)(2) Safe Harbor — strip all 18 identifier classes |
hipaa_expert_review_assist |
redact | redacted; clinical concepts kept | none | optional | Assist Expert Determination (§164.514(b)(1)); keeps microbiology/clinical terms for a statistician to assess residual risk |
gdpr_pseudonymization |
replace | replaced; clinical kept | kept + reversible | mandatory | GDPR Art. 4(5) pseudonymization — reversible under controlled key |
canada_pipeda |
replace (IDs masked) | replaced | kept + reversible | mandatory | PIPEDA-aligned; like GDPR but masks ID_NUM/SSN outright |
research_limited_dataset |
mask direct ids | keeps dates, age, ZIP, geography, org, job | none | mandatory | HIPAA Limited Data Set (§164.514(e)) — usable for research with a DUA |
clinical_minimal_redaction |
mask direct ids | keeps quasi-identifiers | none | optional | Internal clinical use where readability matters; lighter cascade |
strict_no_leak |
mask everything | masked; even clinical concepts masked | none | mandatory | Maximum-recall, union arbitration, all cascade tiers — zero-leakage posture |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 156 lines · 143 tokens per session scan A 2844960e8088
configuring-privacy-policies is a skill published in the GitHub repository maziyarpanahi/openmed (5,302 stars, last pushed today), licensed Apache-2.0. It adds 143 tokens to every session and 2,155 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
benchmarking
Use this skill when the user wants to benchmark an MLX-VLM change and present the numbers in a PR — fork-vs-main A/B comparisons, isolated-module micro-benchmarks, median-of-N timing with warmup, peak-memory reporting, correctness checks, parameter sweeps, and self-contained reproducible bench scripts to paste into a…
convert-quantize
Use this skill when the user wants to convert a Hugging Face model to MLX or quantize/dequantize one with mlxvlm.convert, including bits and group size, quant modes (affine, mxfp4, nvfp4, mxfp8), RTN vs AWQ, mixed-bit recipes, dtype casts, calibration (text or multimodal), local vs Hub paths, revisions, uploading to…
server-inference
Use this skill when the user wants to run or debug MLX-VLM server inference, including uv run mlxvlm.server, /v1/models, /v1/chat/completions, /v1/responses, streaming, OpenAI-compatible clients, health checks, metrics, model unload/reload, adapters, trust-remote-code, and server request/response failures.
cli-inference
Use this skill when the user wants to run or debug MLX-VLM inference from the command line, including uv run mlxvlm.generate, image/audio/video inputs, local model paths, Hugging Face model IDs, deterministic repro commands, and CLI errors around processors, prompts, model loading, or missing weights.
fda-database
Query openFDA API for drugs, devices, adverse events, recalls, regulatory submissions (510k, PMA), substance identification (UNII), for FDA regulatory data analysis and safety research.
meta-compliance-audit-bundle
Auditable compliance bundle: deep-research with citations → signable .docx report → read-only PDF archive → memory note of audit findings.