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.
npx skills add jlintz/claude_uhc_plugins --skill submit-claimgit clone --depth 1 https://github.com/jlintz/claude_uhc_pluginsWrote 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.
[](https://agentmods.dev/skills/jlintz/claude_uhc_plugins/submit-claim)<a href="https://agentmods.dev/skills/jlintz/claude_uhc_plugins/submit-claim"><img src="https://agentmods.dev/badge/skills/jlintz/claude_uhc_plugins/submit-claim.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00060 | $0.05613 |
| Opus 5 | $0.00030 | $0.02806 |
| Sonnet 5 | $0.00012 | $0.01123 |
| Haiku 4.5 | $0.00006 | $0.00561 |
Grade A, and why
submit-claim 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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 366 lines — stays where its author put it; the contents beside it link to each section on GitHub.
UHC Direct Medical Reimbursement — Claim Submission
This skill automates filling out UHC's Direct Medical Reimbursement form at https://memberforms.uhc.com/DirectMedicalReimbursement.html using data extracted from a superbill or medical receipt.
Prerequisites
- Chrome browser open with Chrome DevTools MCP connected
config/member.jsonpopulated with subscriber details including email address (will prompt to create if missing)
When to Use
This skill activates when the user requests to:
- Submit a medical claim or insurance claim
- Fill out the UHC reimbursement form
- Process a superbill or medical receipt for insurance
- Submit an out-of-network claim
Critical: The form lies about whether a field is filled
This is an AEM adaptive form. Every input has two states: the raw DOM input.value, and the framework's bound value that validation actually reads. fill/fill_form can set the first without committing the second. When that happens the field looks correct in a take_snapshot and on screen, but "Next" or "Search" silently does nothing — no error text, no aria-invalid anywhere on the page.
The tell is the wrapper class. input.closest('.guideFieldNode').className must contain validation-success af-field-filled. An uncommitted field stays af-field-empty, usually alongside guideActiveField (meaning it is still focused, so blur never fired).
The most common cause: fill_form leaves its LAST field focused and uncommitted. It blurs each field as it moves to the next, but nothing blurs the final one. In practice, expect to fix the last field of every batch — it is the default case, not the exception. Fields observed hitting this: Group number, provider search ZIP, provider manual-entry ZIP, Charge amount, Provider TIN, added-line Date of service.
The fix (works for every field — TIN, ZIP, dates, amounts) via evaluate_script:
() => {
const n = [...document.querySelectorAll('.guideFieldNode')].find(x => x.className.includes('FIELD_CLASS'));
const el = n.querySelector('input');
const setter = Object.getOwnPropertyDescriptor(window.HTMLInputElement.prototype, 'value').set;
el.focus();
setter.call(el, 'VALUE');
['input','change','blur'].forEach(t => el.dispatchEvent(new Event(t, { bubbles: true })));
el.blur();
return { val: el.value, cls: n.className };
}
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.
- 4d ago Changed · +54 lines 834880a15ea1
- 7d ago First seen · 312 lines · 60 tokens per session scan A de17e8e075d4
submit-claim is a skill published in the GitHub repository jlintz/claude_uhc_plugins (3 stars, last pushed 4d ago), licensed MIT. It adds 60 tokens to every session and 5,613 once invoked, about $0.0003 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 2026-08-31.
Other skills, from other repositories
sector-rotation
An analysis framework for comparing industries in the Chinese A-share stock market, using business conditions, price momentum, valuation, and money flows. It produces rankings and higher- or lower-allocation suggestions.
strategy-pivot-designer
Detect backtest iteration stagnation and generate structurally different strategy pivot proposals when parameter tuning reaches a local optimum.
twitter-reader
Read Twitter/X for financial research using opencli (read-only). Use this skill whenever the user wants to read their Twitter feed, search for financial tweets, view bookmarks, look up user profiles, or gather market sentiment from Twitter/X. Triggers include: "check my feed", "search Twitter for", "show my…
chenhao-limit-up
A framework for judging Chinese A-share stocks that have reached the daily price-rise limit, using market mood, sector leadership, and trading momentum.
trading-risk-gate
Unified pre-trade safety gate: Ruin check (Law #1), ergodicity audit, and win-rate dominance validation. Absorbs: ergodicity-check, law-of-ruin, win-rate-dominance.
vectorbt
High-performance vectorized backtesting with parameter optimization, portfolio simulation, and rich performance metrics.