メニュー
Eval-only routing reference for Tula cross-skill composition tests. Documents which skill handles each patient-agent request type. USE FOR: composition eval tasks that test handoffs between skills. DO NOT USE FOR: end-user production invocations.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | patient-agent-routing |
| description | Eval-only routing reference for Tula cross-skill composition tests. Documents which skill handles each patient-agent request type. USE FOR: composition eval tasks that test handoffs between skills. DO NOT USE FOR: end-user production invocations. |
| metadata | {"openclaw":{"emoji":"🔀"}} |
Eval-only skill loaded by evals/composition/ to test cross-skill coordination.
Aggregates a user's health-related signal feeds (social, web, wearables, portal, calendar) into a curated, scored digest. Personal data is referenced from an external profile file, never embedded. USE FOR: 'run myhealth-pulse', scheduled daily/weekly pulses, mention digests. DO NOT USE FOR: PDFs (med-pdf), clinician messages (epic-note), chart data (health-records), or anything that puts PHI in outbound queries.
Drafts a concise patient-portal message the user can copy-paste to a clinician via Epic MyChart, Oracle Health HealtheLife, or similar. Personal data is referenced from an external profile file, never embedded. USE FOR: symptom reports, refill requests, lab/result follow-ups, visit follow-ups, asking a single clear question. DO NOT USE FOR: clinical notes, SOAP/discharge documentation, insurance or billing letters, or answering the patient's medical question.
Pulls the user's medical records from patient portals (Epic MyChart, Oracle/Cerner) via end-to-end-encrypted SMART on FHIR, into per-provider JSON files Tula can reason over. Personal data is referenced from an external profile file, never embedded. USE FOR: connecting or refreshing records, trending labs across visits, listing meds/conditions/allergies, searching clinical notes. DO NOT USE FOR: parsing a PDF (use med-pdf), drafting clinician messages (use epic-note), or sending PHI outside the workspace.
Personalized ambient health awareness. Pulls environmental, public-health, and place-based data for the user's location and triages it against their longitudinal record, surfacing only what matters. USE FOR: 'lookout for me' briefings, ambient/air/UV/weather/pressure checks, place-based resource queries, travel-mode 'what's [ZIP] like for me'. DO NOT USE FOR: diagnosing, recommending or changing treatment, billing/payor/coding content, or sending PHI to any environmental provider.
Reads medical PDFs (labs, radiology, MyChart/Epic exports, discharge summaries, pathology) and turns them into structured JSON Tula can reason over. Personal data is referenced from an external profile file, never embedded. USE FOR: the user sharing a health-related PDF, image, or screenshot, or asking to compare results across visits. DO NOT USE FOR: non-medical PDFs, generating new clinical reports, or sending PHI outside the workspace.
Surfaces what changed in the user's health since a reference point (date, named event, 'since last X'). Reads workspace memory from med-pdf, health-records, myhealth-pulse, and dated notes; tiers changes by clinical significance. USE FOR: 'what changed', post-visit summaries, longitudinal questions, weekly check-ins. DO NOT USE FOR: news/social (myhealth-pulse), PDFs (med-pdf), chart data (health-records), drafting (epic-note).
