Vellum — AI skill for Claude Code
Vellum: an auditable agentic ICD-10-CM/PCS autocoder that proves every code and catches the model fabricating its own evidence.
How to install Vellum
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open FareedKhan-dev/vellum and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Vellum does
Vellum: an auditable agentic ICD-10-CM/PCS autocoder that proves every code and catches the model fabricating its own evidence. Glass-box web cockpit. Built with Claude.
Alternatives in AI
- Cwc Long Running Agents — Claude Code's built-in /goal command gives you a generator/evaluator loop out of the box: set a completion con 668 ★
- Agentic SEO Skill — An LLM-first SEO analysis skill for Antigravity, Codex, Claude with 16 specialized sub-skills, 10 specialist a 419 ★
- SoloDawn — Catches and repairs AI hallucinations before delivery: three quality gates, 31 built-in rules, a self-healing 312 ★
README
Vellum
Coding you can prove.
**An auditable agentic ICD-10 autocoder whose work you can prove, and the moment it cannot.**
Five Claude Opus roles read a hospital discharge summary and propose the billable codes. Four model-free, network-free engines then prove or refuse every one of them. The point of difference is that Vellum catches the model inventing its own evidence: when the model quotes a sentence that is not in the record, a substring check with no model and no network finds it and holds the code back.
**[▶ Watch the three-minute demo](https://www.youtube.com/watch?v=FBezih99aBI)** · Built with Claude: Life Sciences (Builder track)
For judges
| 🎬 Demo video (3 min) | Vellum: the AI medical coder that catches its own hallucinations |
| 📄 Technical report | PDF · Markdown · [Methodology](METHODOLOGY. |
Related Skills
Metho Agentic
Scaffold AI-powered projects with built-in methodology. Agents, memory, kanban and governance out of the box.
FriesTrader
Robinhood Agentic Trading agent — a fully automated AI trading bot placing real orders through Robinhood's Age
GPT Series Reasoning Style
A delivery-discipline behavior layer for AI agents — turns “the agent says it's done” into “the agent proves i
Retornatus
Repo-native governance for AI coding agents (Cursor, Claude Code, Codex). Agree on the finish line in writing,
AI Debate Club
A multi-agent debate engine. Two different LLMs argue any motion, research the web for themselves, and cite th
Automated Agentic AI Web Agency
Fully Automated Agentic AI Web Agency Built Around Claude Code
Related Agents
Weavie Tester
Proves a change actually works by running the real Weavie app, exercising the scenarios a PR should cover, and
Sf AI Architect
AI Architect of the Software Factory cell. Use whenever a project has an ML/DL/agentic-AI component, to decide
Buy Box Analyst
Fills in what the ranking needs to know about each company. Establishes who actually owns it from filings and