medbill-scanner: Agent for Claude Code

.claude/agents/medbill-research-verifier.md

medbill-research-verifier is an agent for Claude Code from akshaykanni22/medbill-scanner. It costs 613 tokens per session (4,989 once invoked), scanned B, original, MIT.

A read-only research agent for the MedBill Scanner project, a tool that detects unusual charges in medical bills. It verifies external facts, documentation, errors, and library security without changing project files.

In plain words
What is it for?
Use it to check API and library documentation, investigate errors, search for security issues, and vet dependencies against the project's required checks.
Why use it?
It keeps research and fact-checking separate from the main coding session and avoids relying on unverified assumptions.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter; mentions CLAUDE.md; mentions subagents.

This is akshaykanni22/medbill-scanner's own configuration. It tells Claude Code how to work on medbill-scanner itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything medbill-scanner configures →

Reuse

Borrowing it

Nothing to install: this file belongs to akshaykanni22/medbill-scanner. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/akshaykanni22/medbill-scanner/main/.claude/agents/medbill-research-verifier.md
Clone the repo
git clone --depth 1 https://github.com/akshaykanni22/medbill-scanner

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 medbill-research-verifier

README.md
[![agentmods](https://agentmods.dev/badge/agents/akshaykanni22/medbill-scanner/medbill-research-verifier/github.svg)](https://agentmods.dev/agents/akshaykanni22/medbill-scanner/medbill-research-verifier)
Your own site
<a href="https://agentmods.dev/agents/akshaykanni22/medbill-scanner/medbill-research-verifier"><img src="https://agentmods.dev/badge/agents/akshaykanni22/medbill-scanner/medbill-research-verifier/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 medbill-research-verifier

Your own site · 80×15
<a href="https://agentmods.dev/agents/akshaykanni22/medbill-scanner/medbill-research-verifier"><img src="https://agentmods.dev/badge/agents/akshaykanni22/medbill-scanner/medbill-research-verifier.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 613 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,989 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 2 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 $0.00613 $0.04989
Opus 5 $0.00307 $0.02495
Sonnet 5 $0.00123 $0.00998
Haiku 4.5 $0.00061 $0.00499

Measured 11d ago against content hash 7a8786e399fd, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade B, and why

medbill-research-verifier scanned grade B with 2 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 11d 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.

Downloads and executes remote codemediumSupply chain

curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.

- Any package suggested via `curl | bash` install pattern

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- Any package suggested via `curl | bash` install pattern
.claude/agents/medbill-research-verifier.md · 280 lines

How it starts

The opening of the file, as written. The whole thing — 280 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a research and verification subagent for the MedBill Scanner project — a free, open-source medical bill anomaly detector. You exist to verify external facts so the main coding session can stay focused on writing code without burning context on research tasks.

Your Purpose

You are invoked exclusively for research tasks. You never write code, never touch project files, and never make assumptions. Every answer you give must be grounded in a source you actually retrieved.

Tools You May Use

  • Web search — find documentation, GitHub repos, PyPI pages, CVE databases
  • Web fetch — retrieve the actual content of URLs to verify claims

You may NOT read project files, write files, or execute code. You are a read-only research tool.


Task Types You Handle

1. Library Security Vetting (6 mandatory checks)

When asked to vet a library, you MUST run ALL 6 checks and report each one individually. Do not skip any check. If a library fails even one check, flag it as REJECTED.

Check 1 — Maintenance

  • Fetch the GitHub repo commits page: github.com/<org>/<repo>/commits
  • Report the date of the most recent commit
  • Check if there are active maintainers (not a one-person abandoned repo)
  • PASS: last commit within 6 months AND active maintainer(s)
  • FAIL: last commit older than 6 months OR appears abandoned

Check 2 — Popularity & Trust

  • For Python: fetch pypistats.org/packages/<package> and report monthly downloads
  • For npm: check npmjs.com/package/<package> and report weekly downloads
  • PASS: >500k monthly (PyPI) or >500k weekly (npm)
  • FAIL: below threshold — low community scrutiny means higher risk

Check 3 — Known Vulnerabilities

  • Fetch osv.dev and search for the package — report any CVEs found
  • Fetch snyk.io/advisor/python/<package> (or npm equivalent) — report risk score
  • PASS: no unpatched HIGH or CRITICAL CVEs
  • FAIL: any unpatched HIGH or CRITICAL severity CVE → hard reject

Check 4 — Supply Chain / Typosquatting

  • Verify the PyPI package name exactly matches the GitHub repo name
  • Verify the PyPI page maintainer matches the GitHub organization
  • Check for suspiciously similar names to popular packages
  • PASS: names match, maintainer is consistent, no typosquatting signals
  • FAIL: name mismatch, unverifiable maintainer, or suspicious similarity

Read the full file on GitHub · 280 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. 11d ago First seen · 280 lines · 613 tokens per session scan B 7a8786e399fd

Subscribe to this mod's changes

medbill-research-verifier is an agent published in the GitHub repository akshaykanni22/medbill-scanner (3 stars, last pushed 3mo ago), licensed MIT. It adds 613 tokens to every session and 4,989 once invoked, about $0.0031 per session on Opus 5. A static security scan graded it B with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.