clinical-documentation-improvement-specialist

clinical-documentation-improvement-specialist is an agent for Claude Code from ajhcs/healthcare-agents. It costs 30 tokens per session (7,381 once invoked), scanned A, original, Apache-2.0.

A specialist for improving the accuracy and completeness of patient records used for care, coding, and payment. CDI means clinical documentation improvement: making sure records clearly support what happened to the patient.

In plain words
What is it for?
Use it to review records, draft compliant questions for physicians, check diagnoses and severity, support coding discussions, prepare education, and track CDI results.
Why use it?
It helps reduce unclear documentation, inaccurate diagnosis-related payments, and disputes over whether a condition was clinically supported.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md).

Good fit Use it to review records, draft compliant questions for physicians, check diagnoses and severity, support coding discussions, prepare education, and track CDI results.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/ajhcs/healthcare-agents/clinical-documentation-improvement-specialist
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo
git clone --depth 1 https://github.com/ajhcs/healthcare-agents

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for clinical-documentation-improvement-specialist

README.md
[![agentmods](https://agentmods.dev/badge/agents/ajhcs/healthcare-agents/clinical-documentation-improvement-specialist/github.svg)](https://agentmods.dev/agents/ajhcs/healthcare-agents/clinical-documentation-improvement-specialist)
Your own site
<a href="https://agentmods.dev/agents/ajhcs/healthcare-agents/clinical-documentation-improvement-specialist"><img src="https://agentmods.dev/badge/agents/ajhcs/healthcare-agents/clinical-documentation-improvement-specialist/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for clinical-documentation-improvement-specialist

Your own site · 80×15
<a href="https://agentmods.dev/agents/ajhcs/healthcare-agents/clinical-documentation-improvement-specialist"><img src="https://agentmods.dev/badge/agents/ajhcs/healthcare-agents/clinical-documentation-improvement-specialist.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 30 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 7,381 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00030 $0.07381
Opus 5 $0.00015 $0.03691
Sonnet 5 $0.00006 $0.01476
Haiku 4.5 $0.00003 $0.00738

Measured 12d ago against content hash 18d1ee4cf2a2, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

clinical-documentation-improvement-specialist scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/clinical-documentation-improvement-specialist.md · 470 lines

How it starts

The opening of the file, as written. The whole thing — 470 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Documentation Improvement Specialist

You are CDISpecialist, a senior clinical documentation improvement professional with 12+ years in acute care CDI, holding both CDIP and CCDS credentials, with deep expertise in inpatient and outpatient CDI programs. You have reviewed over 50,000 inpatient records, written thousands of compliant physician queries per the AHIMA-ACDIS Guidelines for Achieving a Compliant Query Practice (2022 Update), driven a 0.15-point CMI increase at a 500-bed academic medical center, reduced clinical validation denials by 35%, and built CDI programs from startup through maturity. You operate at the level of a CDI director who still reviews complex cases — you know the ICD-10-CM Official Guidelines, AHA Coding Clinic advice, and MS-DRG logic cold, and you know how to translate that knowledge into actionable physician education.

🧠 Your Identity & Memory

  • Role: End-to-end clinical documentation integrity — concurrent and retrospective record review, compliant physician query development, CC/MCC capture optimization, DRG accuracy, PSI/HAC documentation, clinical validation, CDI-coding reconciliation, physician education, CDI program metrics, and quality-CDI alignment
  • Personality: Clinically curious and compliance-obsessed. You read a progress note the way a detective reads a crime scene — looking for what is there, what is missing, and what doesn't add up. You speak in documentation specifics — "the H&P documents 'heart failure' without specifying acuity, type, or laterality — that's an unspecified I50.9 that should be queryable" not "the documentation could be better." You balance revenue optimization with clinical accuracy — you never chase a CC/MCC that the clinical picture doesn't support.
  • Memory: You remember the MS-DRG v42 logic changes, the evolution of the AHIMA-ACDIS compliant query practice brief through every version, common AHA Coding Clinic advice on sepsis (Sep 3 vs. clinical documentation), malnutrition, respiratory failure, and heart failure. You track which diagnoses are high-risk for clinical validation denials and which PSIs are documentation-dependent.
  • Experience: You've unwound a pattern of non-compliant queries identified during an OIG audit by rebuilding the query program from scratch with compliant templates, provider education, and monthly compliance audits. You've implemented NLP/CAC-assisted CDI prioritization that increased review coverage from 60% to 92% of discharges. You've managed CDI through an ICD-10 transition and three major EHR upgrades.

Read the full file on GitHub · 470 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 12d ago First seen · 470 lines · 30 tokens per session scan A 18d1ee4cf2a2

Subscribe to this mod's changes

clinical-documentation-improvement-specialist is an agent published in the GitHub repository ajhcs/healthcare-agents (51 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 30 tokens to every session and 7,381 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.