Agent skill · operations · jeremylongshore
openevidence-incident-runbook
Triage an OpenEvidence safety, privacy, access, recording, or evidence incident with containment and accountable escalation. Use when working with OpenEvidence in a healthcare organization. Trigger with "openevidence incident runbook", "OpenEvidence incident-response", or a matching workflow request.
Why this skill is useful
Provides specific incident response procedures and triage steps for OpenEvidence that the AI wouldn't reliably generate on its own.
What it needs
About 2k tokens when loaded. Last updated 2026-08-07. 2,604 stars on the source repository.
What this skill does
OpenEvidence Clinical-AI Safety Incident Runbook Overview Protect patients and data first, then preserve a minimum, redacted factual record for clinical and technical resolution. Keep inputs minimal, separate observed facts from assumptions, and leave consequential decisions with the named accountable owner. Prerequisites A clearly bounded workflow, accountable clinical owner, and organizational policy Current first-party OpenEvidence documentation and applicable institution agreements Synthetic or properly authorized minimum-necessary data Tool Discipline Use Read, Glob, and Grep to inspect supplied policies, plans, and evidence. Use WebFetch only for current first-party OpenEvidence documentation. Use Write or Edit only when the user requests a named deliverable with an approved destination. Never expose credentials, PHI, recordings, or unrestricted environment output. Current Contract OpenEvidence output does not replace clinical judgment or emergency procedures. Privacy, recording, and PHI obligations depend on law, policy, consent, and applicable agreements. Security concerns can be reported through OpenEvidence’s published security contact; product support paths govern other issues. Authentication Use only the official OpenEvidence web/mobile sign-in or an institution-approved access path. Do not invent API keys, OAuth clients, SDK credentials, service accounts, or private endpoints. Never ask a user to reveal a password, session token, cookie, or recovery code. Instructions 1. If care may be affected, move to the organization’s clinical escalation or emergency procedure before investigating the tool. 2. Classify patient safety, privacy/PHI, unauthorized access, recording consent, availability, or evidence integrity; set severity. 3. Contain within authority: stop using the affected workflow, prevent further copying, and preserve necessary facts. 4. Notify the accountable clinical, privacy, security, legal, operational, and vendor owners according to policy. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills jeremylongshore/openevidence-incident-runbook