Individual skill
ehr-quality-packet
Build normalized JSON packets from a read-only EHR quality/referral API. Use this skill whenever the user asks for an EHR duplicate-chart merge packet, referral coordination packet, orthopedic/cardiology referral audit, care-transition handoff packet, ServiceRequest quality review, ICD/provider/document evidence reconciliation, or any task that references TASK_ENV_BASE_URL and an answer_template.json schema.
experiments/codex_skill_creator_comparison_gpt5_5_xhigh/reports/skills/task_group_015/fewshot/opencode/fewshot_attempt_02/SKILL.md
Source commit 5082a777379156a048881e94940d640702e5e09a
What was checked
- Matches the observed repository HEAD
- Yes
- Scanner rules are current
- No
- Repository scan complete
- No
- Skill scan complete
- No
- Observed behavior
- Not checked yet
- Last checked
- Nov 6, 2026
These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.
Install this skill
Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.
An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.
Repository context
- Repository stars (not skill ratings)
- 121
- Repository last updated
- Oct 29, 2026
- License
- Apache-2.0
Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.
- agent Repository topic
- ai Repository topic
- benchmark Repository topic
- llm Repository topic
- self-evolving Repository topic
- self-learning Repository topic
- skills Repository topic
- security Derived category
Files and excerpt
File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.
Browse files
Read excerpt
# EHR Quality Packet
Use this skill to produce schema-conformant JSON from the task prompt, local payload templates, and the read-only EHR API. The expected answer is evidence-derived; do not reuse memorized example values.
## Core Workflow
1. Read the user prompt and every local payload file, especially `input/payloads/answer_template.json`.
2. Extract stable identifiers from the prompt and payloads: patient IDs, duplicate candidate IDs, referral IDs, batch IDs, ServiceRequest IDs, provider IDs, service line, and requested packet type.
3. Read `environment_access.md` only to get the base URL and allowed endpoint list. Query only the listed API endpoints.
4. Gather complete evidence before filling the template. Prefer endpoint data over prompt prose when they conflict, except when the template asks for a normalized recommendation rather than the raw record.
5. Fill exactly the top-level keys and field types requested by the template. Return JSON only, with no notes or Markdown.
6. Sort set-like arrays lexicographically unless the template defines a specific order. Preserve special ordering rules such as newest-to-oldest encounters, referral objects by `referral_id`, or risk evidence by `risk_flag`.
Use [`scripts/ehr_collect.py`](scripts/ehr_collect.py) when useful to collect the common evidence bundle: