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 nirholas/three.ws --skill tokenized-agentsgit clone --depth 1 https://github.com/nirholas/three.wsWrote 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/nirholas/three.ws/tokenized-agents)<a href="https://agentmods.dev/skills/nirholas/three.ws/tokenized-agents"><img src="https://agentmods.dev/badge/skills/nirholas/three.ws/tokenized-agents/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/nirholas/three.ws/tokenized-agents"><img src="https://agentmods.dev/badge/skills/nirholas/three.ws/tokenized-agents.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Prompt Injection · line 136 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
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 | $0.00050 | $0.05216 |
| Opus 5.5 | $0.00020 | $0.02086 |
| Sonnet 5.5 | $0.00010 | $0.01043 |
| Haiku 4.5 | $0.00005 | $0.00522 |
Grade A, and why
tokenized-agents 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 — 497 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Before Starting Work
MANDATORY — Do NOT write or modify any code until every item below is answered by the user:
- Agent token mint address (from pump.fun)
- Payment currency decided (USDC or SOL)
- Price/amount confirmed (in smallest unit)
- RPC URL provided or a fallback agreed upon
- Framework confirmed (Next.js, Express, other)
You MUST ask the user for ALL unchecked items in your very first response. Do not assume defaults. Do not proceed until the user has explicitly answered each one.
Safety Rules
- NEVER log, print, or return private keys or secret key material.
- NEVER sign transactions on behalf of a user — you build the instruction, the user signs.
- Always validate that
amount > 0before creating an invoice. - Always ensure
endTime > startTimeand both are valid Unix timestamps. - Use the correct decimal precision for the currency (6 decimals for USDC, 9 for SOL).
- Always verify payments on the server using
validateInvoicePaymentbefore delivering any service. Never trust the client alone — clients can be spoofed. - Always verify your code against this skill before finalizing. Before delivering generated code, re-read the relevant sections of this document and confirm:
- Parameter types match the documented signatures —
buildAcceptPaymentInstructionsacceptsnumberfor numeric fields,validateInvoicePaymentacceptsnumber, andgetInvoiceIdPDAacceptsnumber. - Parameter ordering and names match exactly.
- Default values (e.g.
computeUnitLimitdefaults to100_000) are not contradicted. - Import paths use
@three-ws/agent-payments, not internal module paths.
- Parameter types match the documented signatures —
Supported Currencies
| Currency | Decimals | Smallest unit example |
|---|---|---|
| USDC | 6 | 1000000 = 1 USDC |
| Wrapped SOL | 9 | 1000000000 = 1 SOL |
Environment Variables
Create a .env (or .env.local for Next.js) file with the following:
What ships with it
8 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 · 497 lines · 50 tokens per session scan A 2e26dd3859f9
tokenized-agents is a skill published in the GitHub repository nirholas/three.ws (250 stars, last pushed today), licensed Apache-2.0. It adds 50 tokens to every session and 5,216 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-09-03.
Other skills, from other repositories
blockchain-defi
Use this skill when asked about decentralized finance, DeFi protocols, AMM mechanics, lending and borrowing protocols, perpetual futures, yield farming, liquidity mining, liquid staking, restaking, yield optimization, DeFi security, MEV, and protocol composability. Covers AMM design (Uniswap, Curve, Balancer), lending…
fintech-expert
Expert in financial technology, payment processing, open banking APIs, PSD2, blockchain in finance, robo-advisors, and RegTech. Use when the user mentions payments, open banking, PSD2, blockchain, cryptocurrency, or robo advisor, or when the task involves Payment Processing, Blockchain in Finance, Payment Security, or…
swapper-deposit
Deposit and bridge funds into a wallet or protocol using Swapper Finance. Handles fiat on-ramp (card payments via Mastercard, Visa, Apple Pay, Google Pay), crypto transfers, and cross-chain bridging via Chainlink CCIP — directly into lending, staking, and liquidity protocols. Supports Ethereum, Base, Arbitrum…
stablecoin-chain-explorer
Explore stablecoin TVL distribution and yield opportunities by blockchain. Query which chains have the most stablecoins, compare cross-chain yields, and find the best opportunities on Ethereum, BSC, Arbitrum, Base, Polygon, and more. Also for protocols and issuers researching where stablecoin TVL and liquidity…
crypto-report
Analyze cryptocurrency projects with tokenomics, on-chain metrics, and market analysis. Generate comprehensive crypto research reports.
chord-swap
Swap tokens safely at the best route through Chord (getchord.xyz), a DEX aggregator, on Ethereum, Base, Arbitrum, BNB Chain, Avalanche, Monad, Robinhood Chain (tokenized stocks such as NVDA, TSLA) and Arc, plus Solana. Finds the right token, checks it for honeypots and taxes, quotes the best route, waits for the…