Install
$ agentstack add skill-realactivity-tula-epic-note ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
epic-note
When to Use
✅ Use when:
- The user asks for help drafting a message to their PCP or specialist
- Reporting a new or worsening symptom
- Asking about a recent test or lab result
- Requesting a medication refill
- Following up after a visit, procedure, or referral
When NOT to Use
❌ Don't use when:
- Symptoms suggest an emergency - return the 911 redirect from
[references/triage-rules.md](references/triage-rules.md), don't draft
- Generating a clinician-facing note (SOAP, discharge, dictation)
- Drafting non-clinical letters (insurance, billing)
- The user is asking a medical question for advice - that's answered directly,
not in a portal message
Setup
Profile. Resolve from the precedence in [references/profile-schema.md](references/profile-schema.md#where-the-profile-lives) (skill config -> env var -> ~/.openclaw/workspace/memory/profile.yaml). Use care_team.pcp.display_name as the default greeting recipient when the user doesn't name a clinician. The skill never contains the profile.
Workflow
- Triage first. Scan input for red-flag symptoms. If present, do NOT
draft - return the 911 redirect text from [references/triage-rules.md](references/triage-rules.md).
- Identify the single ask. If multiple unrelated topics exist, propose
separate messages and ask the user which to send first.
- Draft using the canonical format in
[references/portal-message-format.md](references/portal-message-format.md): Subject: line with urgency, greeting, one-sentence ask, 2-4 sentences of context, optional bullet data, sign-off.
- Word budget. Aim for ≤150 words; hard cap 220. Front-load the ask.
Plain language. No self-diagnosis, no filler praise, no PHI of others.
- Output is copy-paste ready. No explanations or meta-commentary
wrapped around the message - just the message itself.
Examples
See [references/examples.md](references/examples.md) for worked input/output pairs (medication side effect, lab follow-up, refill request, multi-topic split).
Privacy
Portal messages travel through health-system networks under HIPAA. Tula generates drafts locally - no PHI leaves the workspace.
- Don't include PHI of family members or third parties.
- The draft stays in the conversation. Never auto-send.
- Don't paste clinician/patient identifiers into web search or external
services for "verification."
Troubleshooting
- No clinician name given -> use
care_team.pcp.display_namefrom the
profile, or Hello [Care Team], if none resolves.
- Vague input -> ask one clarifying question (onset, severity, which
clinician), then draft.
- User asks for medical advice -> decline politely; this skill drafts
messages to clinicians, not answers from them.
- Multi-topic input -> propose separate messages, don't bundle.
- Emergency-flavored prompt -> return triage redirect; never wrap a
drafted message around it.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: realactivity
- Source: realactivity/tula
- License: Apache-2.0
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.