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 understudylabs/understudy-agent-tools --skill inspect-billing-sourcesgit clone --depth 1 https://github.com/understudylabs/understudy-agent-toolsWrote 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/understudylabs/understudy-agent-tools/inspect-billing-sources)<a href="https://agentmods.dev/skills/understudylabs/understudy-agent-tools/inspect-billing-sources"><img src="https://agentmods.dev/badge/skills/understudylabs/understudy-agent-tools/inspect-billing-sources/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.
<a href="https://agentmods.dev/skills/understudylabs/understudy-agent-tools/inspect-billing-sources"><img src="https://agentmods.dev/badge/skills/understudylabs/understudy-agent-tools/inspect-billing-sources.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00000 | $0.01241 |
| Opus 5.5 | $0.00000 | $0.00496 |
| Sonnet 5.5 | $0.00000 | $0.00248 |
| Haiku 4.5 | $0.00000 | $0.00124 |
Grade A, and why
inspect-billing-sources 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 1mo 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 — 143 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Inspect Billing Sources
Use this worker only after the developer asks to use bill evidence, email, an
invoice, a usage export, or an authenticated billing website. This complements
repo inspection: it finds spend hotspots from what the provider charged, then
hands the result back to lower-anthropic-bill.
Safety Gates
- Get explicit approval for the exact read scope before using email, browser, or account surfaces. Name the provider, account, date range, and search/query or page class before reading.
- Prefer exports or screenshots the developer provides. Use Gmail, browser, Chrome, or another connector only when the coding harness exposes that tool and the user approves it in the current thread.
- Read billing metadata only: provider, period, invoice amount, usage amount, model, token counts, cache usage fields, request count, and route labels if already visible. Do not read prompts, completions, traces, API keys, team member lists, customer names, repo names, domains, support tickets, or payment card details.
- Do not send emails, click billing mutations, create API keys, change plans, download private reports, or open raw logs without a separate explicit approval for that exact action.
- Store local artifacts under
.understudy/billing-sources/. Redact or omit account identifiers and invoice IDs unless the developer explicitly requests a private local record.
Intake
Ask for at most one missing input:
- a local usage export, invoice CSV/PDF, or billing screenshot path;
- permission to search email for provider billing messages in a narrow date range;
- permission to inspect a logged-in billing or usage page through the available browser harness.
If no email or browser tool is available, say so and ask for an export or a small manually copied totals table.
Flow
- Resolve supported surfaces. Check the current harness for email and browser tools. If unsupported, use files or manual totals. Do not install a connector just for this path unless the developer explicitly asks.
- Set the scope. Propose the narrowest source: for email, a provider billing query and date range; for browser, one billing/usage page; for files, one export directory. Wait for approval before reading account surfaces.
- Extract observations. Record only billing fields into
.understudy/billing-sources/observations.jsonusing this shape:
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago First seen · 143 lines · 0 tokens per session scan A dea2aaaee47d
inspect-billing-sources is a skill published in the GitHub repository understudylabs/understudy-agent-tools (18 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,241 tokens. 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.
Other skills, from other repositories
x402
Set up Browser Use Cloud payments with x402 — pay per request from a crypto wallet (USDC on Base mainnet), no signup or API key. Two setups it works out up front — "just use it" (set up a wallet so you or Claude Code can run cloud browser tasks paid from the wallet — Claude writes and runs throwaway scripts, nothing…
stock-api
Fetch real-time stock quotes, K-line (candlestick) history, and search symbols for China A-shares, Hong Kong, and US markets. Use when the user asks for a stock price, a quote, market data, candlestick/K-line data, or to look up a stock code by company name.
american-airlines
Check American Airlines AAdvantage balance, elite status, and loyalty points via Patchright. Handles email 2FA with 6-box code entry. Uses persistent browser profiles to skip 2FA on subsequent runs.
vfx
Build real-time 2D effects in-engine — explosions, hit sparks, trails, smoke, pickups, shockwaves, weather — with particle recipes, readability rules and mobile budgets.
israeli-freelancer-ops
Manage daily operations for Israeli freelancers (osek murshe, osek patur) - invoice aging with collection reminders, utility bill collection via browser automation, tax deadline alerts (VAT, Bituach Leumi, mkdamot, annual report), osek patur threshold monitoring, and organized accountant packages (havila). Use when a…
find-my-bookings-and-balances
Uses your own email and browser connectors together to find booking confirmations and points balances, then saves them to Award Travel Finder. Use when the user asks to find, sync or set up all their trips and balances and an email connector (Gmail, Outlook, Apple Mail) or browser control (Claude in Chrome, computer…