Agent skill · data analytics · jeremylongshore
openevidence-performance-tuning
Improve OpenEvidence question quality and reviewer efficiency through controlled prompt refinement. Use when working with OpenEvidence in a healthcare organization. Trigger with "openevidence performance tuning", "OpenEvidence prompting", or a matching workflow request.
Why this skill is useful
Provides specific caching strategies, batch operations, and rate limit management patterns for optimizing OpenEvidence API performance 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 Prompt Refinement Overview Tune context and question structure while holding clinical accountability and source review constant. 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 The official guide publishes prompt guidance and dedicated workflows for complex cases and Snow. More detail is not always safer; include only relevant, authorized context. Performance means decision usefulness and evidence traceability, not fastest answer or longest response. 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. State the decision, intended user, population, outcome, constraints, and what uncertainty must remain visible. 2. Remove identifiers and irrelevant narrative; separate known facts from assumptions. 3. Run a baseline synthetic or authorized de-identified question and score relevance, citations, applicability, and reviewer effort. 4. Change one prompt element at a time: specificity, timeframe, comparator, output structure, or request for conflicting evidence. 5. Open citations and have a qualified clinician compare versions using the same rubric. 6. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills jeremylongshore/openevidence-performance-tuning--86a317