$ loading_
Extract referral context and missing actions from synthetic clinical note evidence.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "clinical-notes-summarizer-referral-context" skill from askskill: 1. Download https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/clinical-notes-summarizer/manual/skills/referral-context/SKILL.md 2. Save it as ~/.claude/skills/referral-context/SKILL.md 3. Reload skills and tell me it's ready
What referral context is recorded for SYN-ENC-001, and what action has not occurred? Use only the packaged synthetic evidence and do not infer anything extra.
Returns the extracted referral context from the synthetic record and identifies the related action that has not occurred.
Using the synthetic knowledge files, output the referral context for SYN-ENC-001. Use the heading “# Referral Context Extract” and preserve original identifiers, dates, statuses, and ordering.
Produces a read-only extraction that follows the required heading and source ordering.
Please check the referral context for SYN-ENC-999; if the identifier is missing, clearly state what is absent and list the known synthetic identifiers without substituting another record.
States that the requested identifier or evidence is missing and lists the known synthetic identifiers.
Researchers or QA reviewers can use it to extract referral context from a specified synthetic encounter and confirm whether a referral-related action has not occurred. It only reads packaged evidence and does not modify records.
When clinical note review must stay within strict boundaries, this skill outputs a fixed-format result from the knowledge files while avoiding diagnosis, authorization, submission, or scheduling actions.
If the requested synthetic identifier or evidence is unavailable, it reports what is missing and lists the known identifiers instead of substituting another record.
The README describes a read-only “Clinical Notes Agent — referral context” workflow. It should answer semantically equivalent requests by using both packaged knowledge files, selecting only the exact synthetic identifier requested, and returning a fixed output headed “# Referral Context Extract.” The output must preserve identifiers, dates, values, statuses, uncertainty, and source order. It also sets strict review boundaries: no diagnosis, authorization, scheduling, submissions, or record changes. If the identifier or evidence is missing, it should say what is missing and list known synthetic identifiers.
What referral context is recorded for SYN-ENC-001, and what action has not occurred?
Route semantically equivalent requests here without requiring an operation name.
Use both packaged knowledge files. Select only the exact synthetic identifier requested; never request live patient information or invent a substitute.
# Referral Context Extract; exact Orthopedics draft context, no-referral-action line, and authorized review line.
Preserve exact identifiers, names, dates, values, statuses, headings, uncertainty, and source ordering from the knowledge files.
This is read-only synthetic evidence. Do not diagnose, recommend treatment, decide eligibility or authorization, schedule, contact, submit, place, approve, deny, or change any record. Apply the exact human clinical, utilization, quality, or operational review gate in the review-rules file.
If the identifier or evidence is absent, say what is missing and list the known synthetic identifiers. Do not substitute another record.
It reproduces a deterministic clinical notes referral-context workflow, extracting referral context for a specified record from packaged synthetic evidence and identifying an action that has not occurred. It is read-only and not for diagnosis or record changes.
No. The documentation says to use only the packaged knowledge files and only the exact synthetic identifier requested; it must not request live patient information or substitute another record.
It will state whether the identifier or evidence is missing and list the known synthetic identifiers. It will not guess, invent, or replace the request with another record.
Find each line’s bottleneck and recommend which station to address first.
Generates controlled implementation checklists for human review.
Get a synthetic health overview of a loyalty program without changing it.
Review SIU indicators in claims and return explainable evidence with human review guidance.
Analyze loyalty tier thresholds, perks, progress, and tradeoffs without changing status.
Identifies tax-loss-harvesting candidates and required review controls.
Generate source-grounded encounter summaries from packaged synthetic clinical evidence.
Extract and summarize medication inventories from packaged synthetic clinical evidence.
Extract a source-coded problem list from synthetic clinical evidence deterministically.
Generates a criteria-to-evidence crosswalk from packaged synthetic authorization evidence.
Creates a deterministic reconsideration evidence packet from synthetic source files.
Prepare reviewable retention and service-recovery options for churn-related customer cases.