Bitterbot is a local-first personal AI agent that runs on a user’s devices, keeps persistent memories, performs tasks, and can exchange reusable skills with other agents. It is intended for people who want a personal assistant that remains available across conversations and activities. The catalogue entries provide instructions and agents for working with Bitterbot.
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 Bitterbot-AI/bitterbot-desktop --skill wallet-paymentsgit clone --depth 1 https://github.com/Bitterbot-AI/bitterbot-desktopWrote 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/bitterbot-ai/bitterbot-desktop/wallet-payments)<a href="https://agentmods.dev/skills/bitterbot-ai/bitterbot-desktop/wallet-payments"><img src="https://agentmods.dev/badge/skills/bitterbot-ai/bitterbot-desktop/wallet-payments/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/bitterbot-ai/bitterbot-desktop/wallet-payments"><img src="https://agentmods.dev/badge/skills/bitterbot-ai/bitterbot-desktop/wallet-payments.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00057 | $0.01134 |
| Opus 5.5 | $0.00023 | $0.00454 |
| Sonnet 5.5 | $0.00011 | $0.00227 |
| Haiku 4.5 | $0.00006 | $0.00113 |
Grade A, and why
wallet-payments 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 16d 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 — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Wallet payments (USDC on Base)
You have a Coinbase Smart Wallet on Base loaded with USDC. Gas is sponsored by the Coinbase Paymaster (zero ETH needed) and USDC on Base has near-zero transaction fees, so micropayments are viable.
Wallet actions
get_balance(token="USDC"): check your USDC balance before making payments.get_address: the wallet address (for the user to fund externally).pay_for_resource: pay for a paywalled HTTP resource via the x402 protocol. Signs the payment AND fetches the content in one call. Requiresresource_urlandamount; optionalreason.fund_wallet: a URL for the user to fund the wallet (Coinbase Onramp on mainnet, faucet on testnet).send_usdc: send USDC to an address for user-initiated transfers, paying other agents or services, purchasing digital goods, or any prompt-driven payment.get_transaction_history: recent wallet transactions.
Approval
Sends and payments are held for the owner's approval unless a standing spend grant covers them. When a wallet call returns APPROVAL-REQUIRED or APPROVAL-PENDING with an id like rv-1a2b3c4d: the action was NOT performed, the owner has been asked, and the gateway will carry it out for them if they approve. Tell the user it is waiting for their approval, do not retry the call, and do not try another way to make the same payment. You will be told the outcome in a later turn.
Handling paywalls (HTTP 402)
When web_fetch returns a 402 Payment Required response, follow this workflow exactly:
- Extract the price: if the 402 result has
price_usdc, use it. The tool derives it from the x402accepts[]offer only when the asset is USDC;x402_offernames the asset, network and amount in atomic units. Otherwise look atx402_headersforx-payment-amount. Do not convert rawamountormaxAmountRequiredvalues frompayment_required,payment_detailsor thebody_snippetyourself: they are atomic units (10000is 0.01 USDC) and easy to misread, so treat the price as unclear. - Handle unclear prices: if you cannot confidently determine the price, DO NOT GUESS. Ask: "This resource requires payment, but the price isn't clear. Would you like me to proceed, and what is your maximum budget?"
- Check balance and rules: call
get_balance(token="USDC"). If the cost exceeds your spending limits, ask the user for permission. - Inform the user: state the cost and your intent briefly (for example "This endpoint requires a 0.15 USDC micropayment. I have the funds and am purchasing it now.").
- Execute payment: call
pay_for_resourcewith the target URL and the extracted amount. - Process content directly:
pay_for_resourcereturns the un-paywalled content. Do NOT callweb_fetchagain after paying; use thepay_for_resourceresult immediately.
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.
- 16d ago Changed eba897d69a5b
- 1mo ago Changed · +4 lines 58fce7ff3afb
- 1mo ago First seen · 53 lines · 57 tokens per session scan A cc9673fe212a
wallet-payments is a skill published in the GitHub repository Bitterbot-AI/bitterbot-desktop (2,481 stars, last pushed yesterday), licensed MIT. It adds 57 tokens to every session and 1,134 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 2026-09-21.
wallet-payments compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| wallet-paymentsBitterbot-AI/bitterbot-desktop | Skill | 2.5k | yesterday | MIT | A | 57 | Claude Code, Codex |
| t2000-jobmission69b/t2000 | Skill | 23 | 1mo ago | MIT | A | 164 | Claude Code |
| t2000-receivemission69b/t2000 | Skill | 23 | 1mo ago | MIT | A | 71 | Claude Code |
| t2000-sendmission69b/t2000 | Skill | 23 | 1mo ago | MIT | A | 83 | Claude Code |
| t2000-swapmission69b/t2000 | Skill | 23 | 1mo ago | MIT | A | 93 | Claude Code |
| t2000-financial-amountsmission69b/t2000 | Skill | 23 | 1mo ago | MIT | A | 106 | Claude Code |
| t2000-check-balancemission69b/t2000 | Skill | 23 | 1mo ago | MIT | A | 61 | Claude Code |