agentleFS
nathan8823nathan8823/fairbill/.claude/skills/review-bill/SKILL.md

review-bill

Skill1 starsChanged 4 months agoMITView on GitHub

Audit an itemized medical bill for errors — duplicates, upcoding, unbundling, phantom charges. Use once an itemized bill with CPT codes is in the case folder.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Transcribe: every line into a table in case.md: date, CPT/HCPCS, description, units, charge.

    line 14
  2. Distinctive:

    If insured: EOB mismatch — any line where the provider's ask exceeds the EOB's patient responsibility

    line 41
  3. Distinctive:

    Never assert fraud — assert "these charges appear inconsistent with the services received" and request correction and a corrected statement.

    line 54

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Review the itemized bill

Prereq: an itemized statement with CPT/HCPCS codes in the case folder. If there isn't one, send the request first (templates/letters/itemized-request.md) — never audit a summary bill.

Steps

  1. Transcribe every line into a table in case.md: date, CPT/HCPCS, description, units, charge. Reading the bill into structured data IS the audit's foundation — most errors surface here.
  2. Decode the codes. For each CPT, state in plain English what it is (and its E/M level if it's a visit code). Flag anything that doesn't match the story from intake. The user was there — walk the timeline with them: "You're billed for two chest X-rays — do you remember two?"
  3. Run the error checklist (details in knowledge/tactics.md §3 and knowledge/glossary.md). Billing departments have insider names for the big four — use them; they signal you know the game:
    • "Double dip" — duplicates (same code + date, or same service under two revenue codes)
    • "Level creep" — upcoding (ER E/M level vs. what happened; observation billed as admission)
    • "Unbundling" — components billed alongside their bundle code
    • "Phantom charges" — drugs/supplies/tests not received; ordered then canceled but billed
    • Quantity errors (units column!) and impossible times (OR minutes, room days — the discharge-day room charge is a classic)
    • Room-type charges: private-room rates for shared rooms. The hospital's bed-management system logs every room/bed assignment by shift (ADT — admission/discharge/transfer records). Request them for any inpatient stay with room charges — see the optional items in templates/letters/records-request.md.
    • Wrong patient, date, or provider lines
    • If insured: EOB mismatch — any line where the provider's ask exceeds the EOB's patient responsibility
  4. Score each finding: confirmed error / suspicious (needs their records) / fine. For "suspicious," the follow-up is a HIPAA records request (templates/letters/records-request.md) — medical records vs. bill is the definitive comparison.
  5. Write the dispute. Confirmed + suspicious items go into templates/letters/error-dispute.md, itemized with codes, one paragraph per item, and land in letters/ for user approval. Disputed amount is frozen; undisputed remainder proceeds to benchmark-price → strategy.

Never assert fraud — assert "these charges appear inconsistent with the services received" and request correction and a corrected statement.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.