aks129/HealthClawGuardrails/skills/fasten-connect/SKILL.md
fasten-connect
Use this skill whenever connecting a patient's real health records from EHR systems (Epic, Cerner, Athena) or the TEFCA national network into HealthClaw Guardrails. Covers: Fasten Stitch widget embed, org_connection_id registration, EHI export job tracking, NDJSON ingestion status, TEFCA IAS identity-verified multi-provider retrieval, and post-import Curatr quality scan workflow.
- Source repository stars
- 27
- Declared platforms
- 0
- Static risk flags
- 3
- Last source update
- 2026-08-04
- Source checked
- 2026-08-04
Decision brief
What it does—and where it fits
Connects patient-authorized real EHR data into HealthClaw Guardrails via Fasten Connect. Once ingested, all records flow through the full guardrail stack: PHI redaction, audit trail, step-up authorization, and tenant isolation.
Not for
- Background download uses a daemon thread — if the process restarts mid-download,
- Download links expire after 24 hours — webhook triggers immediate download
Compatibility matrix
Platform support, with evidence labels
| Platform | Status | Evidence | What to check |
|---|---|---|---|
| Codex | Not declared | No explicit evidence | Portability before use |
| Claude Code | Not declared | No explicit evidence | Portability before use |
| Cursor | Not declared | No explicit evidence | Portability before use |
| Gemini CLI | Not declared | No explicit evidence | Portability before use |
Installation
Inspect first. Install second.
The source command is displayed only when detected. A safe inspection prompt is always available so your agent can explain every action before execution.
npx skills add https://github.com/aks129/HealthClawGuardrails --skill "skills/fasten-connect"Inspect the Agent Skill "fasten-connect" from https://github.com/aks129/HealthClawGuardrails/blob/b94ae9ea95695d5755e22b02ef7b7385f0e3aee7/skills/fasten-connect/SKILL.md at commit b94ae9ea95695d5755e22b02ef7b7385f0e3aee7. List every install step, command, network request, credential, file read/write, external action, and rollback step. Explain whether it fits my task. Do not install or execute anything until I approve.
Workflow
What the source asks the agent to do
- 01
CLEAR verification (test mode)
Phone: (from Fasten test patient credentials page) Email: (from Fasten test patient credentials page)
Phone: (from Fasten test patient credentials page) Email: (from Fasten test patient credentials page) - 02
ID.me verification (test mode)
Username: (test patient email) Password: IDme2026!!
Username: (test patient email) Password: IDme2026!! - 03
Post-Import Curatr Workflow
After ingestion completes, run Curatr evaluation on clinical resources:
Deprecated code systems (ICD-9-CM flagged as critical)Invalid RxNorm, LOINC, SNOMED, CVX codesMissing US Core required fields (clinicalStatus, verificationStatus, etc.) - 04
Setup
Review the “Setup” section in the pinned source before continuing.
Review and apply the “Setup” source section. - 05
Environment Variables
Review the “Environment Variables” section in the pinned source before continuing.
Review and apply the “Environment Variables” source section.
Permission review
Static risk signals and limitations
Network access
The documentation includes network, browsing, or remote request actions.
<link rel="stylesheet" href="https://stitch.fastenhealth.com/v0.4/bundle.css">Network access
The documentation includes network, browsing, or remote request actions.
<script src="https://stitch.fastenhealth.com/v0.4/bundle.js"></script>Sends data out
The documentation includes sending, uploading, or posting data to a remote service.
curl -X POST https://api.connect.fastenhealth.com/v1/bridge/fhir/ehi-export \Runs scripts
The documentation asks the agent to run terminal commands or scripts.
docker-compose up -d --buildEvidence record
Why each signal appears
| Signal | Value | Evidence type | Meaning |
|---|---|---|---|
| Quality score | 86/100 | Computed | Documentation, specificity, maintenance, and trust rules |
| Repository stars | 27 | Source | Repository attention, not individual Skill quality |
| Compatibility | 0 platforms | Source | Declared in the catalog source record |
| Usage guide | automated source guide | Editorial | Generated or reviewed according to the visible evidence level |
Pinned source
Provenance and original SKILL.md
- Repository
- aks129/HealthClawGuardrails
- Skill path
- skills/fasten-connect/SKILL.md
- Commit
- b94ae9ea95695d5755e22b02ef7b7385f0e3aee7
- License
- MIT
- Collected
- 2026-08-04
- Default branch
- main
View the original SKILL.md
Fasten Connect — Patient Health Record Ingestion
Connects patient-authorized real EHR data into HealthClaw Guardrails via Fasten Connect. Once ingested, all records flow through the full guardrail stack: PHI redaction, audit trail, step-up authorization, and tenant isolation.
Two modes:
- Standard — patient authenticates with their EHR portal (Epic, Cerner, Athena, etc.)
- TEFCA IAS — single identity verification (CLEAR / ID.me) retrieves records from all QHINs the patient has records at, without per-provider logins
Environment Variables
| Variable | Required | Description |
|---|---|---|
FASTEN_PUBLIC_KEY | Yes | Fasten API public key (public_test_* or public_live_*) |
FASTEN_PRIVATE_KEY | Yes | Fasten API private key — never expose client-side |
FASTEN_WEBHOOK_SECRET | Recommended | Standard-Webhooks HMAC secret from Fasten portal |
FASTEN_CURATR_SCAN | No | Set true to run Curatr quality scan after each import |
Integration Flow
1. Embed the Fasten Stitch Widget
<!-- Standard mode -->
<fasten-stitch-element public-id="your_public_test_key"></fasten-stitch-element>
<!-- TEFCA IAS mode — identity-verified, multi-provider -->
<fasten-stitch-element public-id="your_public_test_key" tefca-mode="true"></fasten-stitch-element>
<link rel="stylesheet" href="https://stitch.fastenhealth.com/v0.4/bundle.css">
<script src="https://stitch.fastenhealth.com/v0.4/bundle.js"></script>
2. Handle the Widget Callback
document.querySelector('fasten-stitch-element')
.addEventListener('widget.complete', async (event) => {
const { org_connection_id, endpoint_id, tefca_directory_id,
platform_type, connection_status } = event.detail;
// Register connection with HealthClaw Guardrails
await fetch('/fasten/connections', {
method: 'POST',
headers: {
'Content-Type': 'application/json',
'X-Tenant-Id': YOUR_TENANT_ID,
},
body: JSON.stringify({
org_connection_id,
endpoint_id,
tefca_directory_id, // use this as stable ID in TEFCA mode
platform_type,
connection_status,
}),
});
});
3. Configure Your Webhook
In the Fasten Developer Portal, set your webhook URL:
https://your-domain.com/fasten/webhook
The webhook fires patient.ehi_export_success when the export is ready.
HealthClaw automatically downloads and ingests the FHIR NDJSON files.
4. Trigger an EHI Export (optional — Fasten may auto-trigger)
curl -X POST https://api.connect.fastenhealth.com/v1/bridge/fhir/ehi-export \
-u "public_test_XXX:private_test_XXX" \
-H "Content-Type: application/json" \
-d '{"org_connection_id": "your-org-connection-id"}'
5. Monitor Ingestion
# List jobs for your tenant
curl /fasten/jobs -H "X-Tenant-Id: tenant-001"
# Poll a specific job
curl /fasten/jobs/<task_id> -H "X-Tenant-Id: tenant-001"
Job status lifecycle: pending → downloading → ingesting → complete | failed
HealthClaw API Endpoints
| Method | Path | Auth | Description |
|---|---|---|---|
POST | /fasten/webhook | HMAC (Fasten-signed) | Receive Fasten events |
POST | /fasten/connections | X-Tenant-Id | Register connection after Stitch widget |
GET | /fasten/connections/<id> | X-Tenant-Id | Connection status |
GET | /fasten/jobs | X-Tenant-Id | List ingestion jobs |
GET | /fasten/jobs/<task_id> | X-Tenant-Id | Single job status |
Guardrails Applied to Ingested Data
Fasten imports pass through the full HealthClaw guardrail stack. Once a resource is ingested, all existing guardrails apply automatically:
| Guardrail | Behavior |
|---|---|
| PHI redaction | Applied on every read — names to initials, identifiers masked, addresses stripped |
| Audit trail | Every resource write creates an immutable AuditEvent |
| Tenant isolation | org_connection_id is bound to one tenant_id at registration |
| Curatr scan | Optional: runs quality evaluation post-import (FASTEN_CURATR_SCAN=true) |
TEFCA IAS Specifics
- Enable with
tefca-mode="true"on the Stitch widget - Use
tefca_directory_id(notendpoint_id) as the stable identifier - Scope is always
patient/*.read— no narrower scope negotiation - Identity verification via CLEAR (phone + email) or ID.me (username/password)
- May return
tefca_no_documents_foundwhen health systems have no records - Additional live-mode fees apply — test with synthetic patients first
TEFCA Test Patients
# CLEAR verification (test mode)
Phone: (from Fasten test patient credentials page)
Email: (from Fasten test patient credentials page)
# ID.me verification (test mode)
Username: (test patient email)
Password: IDme2026!!
# CCDA fixture override (API-mode testing)
POST /v1/bridge/fhir/ehi-export
{"org_connection_id": "...", "fixtures": {"tefca_ccda": "myra-jones.xml"}}
Post-Import Curatr Workflow
After ingestion completes, run Curatr evaluation on clinical resources:
1. GET /fasten/jobs/<task_id> → wait for status: "complete"
2. fhir.search(patient, Condition) → list ingested Conditions
3. curatr.evaluate(Condition, id) → check each for coding issues
4. Present issues to patient in plain language
5. Patient approves fixes
6. curatr.apply_fix(resource, fixes) → apply with Provenance trail
Curatr checks ingested resources for:
- Deprecated code systems (ICD-9-CM flagged as critical)
- Invalid RxNorm, LOINC, SNOMED, CVX codes
- Missing US Core required fields (clinicalStatus, verificationStatus, etc.)
Webhook Events
| Event | Trigger | Action |
|---|---|---|
patient.ehi_export_success | Export ready | Auto-download + ingest |
patient.ehi_export_failed | Export failed | Record failure reason |
patient.authorization_revoked | Consent expired | Mark connection revoked |
patient.connection_success | Widget complete | Optional auto-register |
webhook.test | Config test | Accepted silently |
Note: Never log raw webhook payloads — patient.request_support events
explicitly may contain PHI per Fasten documentation.
File Size Handling
Fasten exports range from ~30MB to 3GB+. HealthClaw uses HTTP streaming
(httpx stream()) to handle large files without loading them into memory.
Progress is committed to the database every 50 resources.
Vercel limitation: Serverless functions timeout before large downloads complete.
Railway or any persistent server deployment is required for production use with
Fasten Connect. See railway.toml for Railway deployment config.
Setup
# Local development
export FASTEN_PUBLIC_KEY=public_test_XXX
export FASTEN_PRIVATE_KEY=private_test_XXX
export FASTEN_WEBHOOK_SECRET=whsec_XXX # from Fasten Developer Portal
export FASTEN_CURATR_SCAN=true
docker-compose up -d --build
# Test webhook locally (use Fasten webhook simulator or ngrok)
# Fasten Webhook Simulator: https://docs.connect.fastenhealth.com/guides/webhook-debugging-simulator.md
Known Limitations
- Background download uses a daemon thread — if the process restarts mid-download,
the job remains in
downloadingstatus (re-trigger by calling the Fasten API again) - Download links expire after 24 hours — webhook triggers immediate download
- TEFCA live mode has additional per-use fees (test mode is free with synthetic patients)
- Upstream tenant isolation is local only — Fasten does not enforce tenant boundaries
Alternatives
Compare before choosing
mgiovani/cc-arsenal
team-review
Multi-agent review team: architecture, security, performance, testing, style, docs/UX, plus an adversary that cross-examines the other 6, for security-sensitive, architectural, or large PRs (15+ files) where a single-agent pass risks missing cross-cutting issues. Use for auth/payments/PII changes, schema/pattern changes, compliance sign-off, or when asked to 'get the review team on this' / 'multi-agent review' / 'thorough review before merge'. For a standard PR or a quick pre-merge check, use /r
dotnet/skills
dotnet-webapi
Guides creation and modification of ASP.NET Core Web API endpoints with correct HTTP semantics, OpenAPI metadata, and error handling. USE FOR: adding new API endpoints (controllers or minimal APIs), wiring up OpenAPI/Swagger, creating .http test files, setting up global error handling middleware. DO NOT USE FOR: general C# coding style, EF Core data access or query optimization (use optimizing-ef-core-queries), frontend/Blazor work, gRPC services, or SignalR hubs.
fcakyon/claude-codex-settings
livekit-skills
Build voice AI agents with LiveKit Agents SDK. Use when the user asks to "build a voice agent", "create a LiveKit agent", "add voice AI", "implement handoffs", "structure agent workflows", or is working with LiveKit Agents SDK. Covers both LiveKit Cloud and self-hosted deployments using lk CLI.
PramodDutta/qaskills
API Test Suite Generator
Automatically generate comprehensive API test suites from OpenAPI specifications covering CRUD operations, error handling, authentication, pagination, and edge cases