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 enforcing-nophi-logginggit 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/enforcing-nophi-logging)<a href="https://agentmods.dev/skills/maziyarpanahi/openmed/enforcing-nophi-logging"><img src="https://agentmods.dev/badge/skills/maziyarpanahi/openmed/enforcing-nophi-logging/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/enforcing-nophi-logging"><img src="https://agentmods.dev/badge/skills/maziyarpanahi/openmed/enforcing-nophi-logging.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 | $0.00139 | $0.01892 |
| Opus 5 | $0.00069 | $0.00946 |
| Sonnet 5 | $0.00028 | $0.00378 |
| Haiku 4.5 | $0.00014 | $0.00189 |
Grade A, and why
enforcing-nophi-logging 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 — 164 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Enforcing No-PHI Logging
Logs are a top breach vector: a clinical string lands in a log line, gets shipped to a centralized log store and an error tracker, and is now PHI sitting outside the de-id boundary. OpenMed's local-first stance says no raw PHI in logs, caches, or error reports — this skill enforces it with a redaction guard that runs before any record is emitted.
When to use this skill
- An OpenMed service logs request text, model output, or exception messages.
- You ship logs/traces to a centralized store or error tracker (Sentry, ELK).
- You need a
logging.Filter(or OTel processor) that redacts PHI pre-emit. - You want structured, no-PHI log fields (offsets, hashes, counts) for debugging.
Quick start — a redacting logging.Filter
import logging
import re
import openmed
# Cheap regex pre-filter for the highest-risk structured identifiers. This runs
# on every record, so keep it fast; the model is the fallback for free-text PHI.
_FAST_PATTERNS = [
(re.compile(r"\b\d{3}-\d{2}-\d{4}\b"), "[SSN]"),
(re.compile(r"\b\d{16}\b"), "[CARD]"),
(re.compile(r"\b[\w.+-]+@[\w-]+\.[\w.-]+\b"), "[EMAIL]"),
(re.compile(r"\b(?:\+?\d[\d().\-\s]{7,}\d)\b"), "[PHONE]"),
]
class NoPHIFilter(logging.Filter):
"""Redact PHI from a log record before it is emitted. Fail closed."""
def __init__(self, model_name: str | None = None, use_model: bool = True):
super().__init__()
self.model_name = model_name
self.use_model = use_model
def filter(self, record: logging.LogRecord) -> bool:
try:
message = record.getMessage()
record.msg = self._scrub(message)
record.args = () # message already rendered & scrubbed
except Exception:
# Never let the logger leak on error — drop the message, keep the level.
record.msg = "[REDACTED: scrub error]"
record.args = ()
return True # keep the (now-clean) record
def _scrub(self, text: str) -> str:
for pattern, tag in _FAST_PATTERNS:
text = pattern.sub(tag, text)
if not self.use_model:
return text
# Model fallback for free-text PHI (names, locations, dates). Replace by
# offset, right-to-left, so earlier offsets stay valid.
spans = openmed.extract_pii(text, model_name=self.model_name) \
if self.model_name else openmed.extract_pii(text)
for e in sorted(spans.entities, key=lambda s: s.start, reverse=True):
text = text[:e.start] + f"[{e.label}]" + text[e.end:]
return text
# Attach to every handler that might emit clinical text.
handler = logging.StreamHandler()
handler.addFilter(NoPHIFilter(model_name="OpenMed/Privacy-PII-Detection"))
logging.getLogger("openmed.service").addHandler(handler)
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 · 164 lines · 139 tokens per session scan A d2f1a530a212
enforcing-nophi-logging is a skill published in the GitHub repository maziyarpanahi/openmed (5,290 stars, last pushed yesterday), licensed Apache-2.0. It adds 139 tokens to every session and 1,892 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
add-new-model
Use this skill when the user wants to add or port a new model architecture to MLX-VLM — mapping a Hugging Face modeltype to a new file under mlxvlm/models, writing the ModelConfig, matching layer/weight names, reusing a similar existing model, adding a test class, and validating the port. Covers vision-language…
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.
contributing
Use this skill when the user wants to contribute to MLX-VLM — opening a PR, where model code/config/tests go, backward-compatible config args, running the test suite, code formatting and the pre-commit hooks (black, clang-format), and PR expectations (tests, review, perf evidence). Use it to set up a change so it…