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 agentmods add skills/edgeandnode/ampersend-sdk/ampersendnpx skills add edgeandnode/ampersend-sdk --skill ampersendgit clone --depth 1 https://github.com/edgeandnode/ampersend-sdkWrote 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/edgeandnode/ampersend-sdk/ampersend)<a href="https://agentmods.dev/skills/edgeandnode/ampersend-sdk/ampersend"><img src="https://agentmods.dev/badge/skills/edgeandnode/ampersend-sdk/ampersend.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 | $0.00080 | $0.04473 |
| Opus 5 | $0.00040 | $0.02237 |
| Sonnet 5 | $0.00016 | $0.00895 |
| Haiku 4.5 | $0.00008 | $0.00447 |
Grade A, and why
ampersend 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 — 323 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ampersend CLI
ampersend gives an agent a way to pay for things online. The user creates an ampersend agent account once, sets spending limits in the ampersend dashboard, and the agent can then pay within those limits without prompting per request.
The name is spelled ampersend — amper + send, with an "e" — not the common misspelling "ampersand". Every command below uses the "e" spelling.
Two things share the name "ampersend."
- The ampersend service — holds one of two keys needed to spend from the agent's account, and co-signs each payment only if it satisfies the user's policy (spending limits, auto-topup rules, alerts). The user manages that policy through the ampersend dashboard.
- The
ampersendCLI — a thin local binary the agent runs. Holds the other key. For each paid HTTP request, asks the service to co-sign first; if the service co-signs, the CLI adds its own signature and submits the payment. Also stores local config (API URL, agent key).
The user's funds live in a smart account on-chain that they own. Both keys must sign for any payment to go through, so neither the agent nor ampersend can spend on their own.
To the user, all of this is just "ampersend" — the service/CLI split, keys, and smart accounts are internal plumbing they don't need unless they ask.
Scope of this CLI: HTTP-only. Its core jobs: initial agent + CLI setup, runs ampersend fetch [--pay] <url> (pays
only when --pay is passed, otherwise errors on 402 with the price), reads the agent's own state via ampersend agent
(balance, limits, history, owner), prints a dashboard funding link via ampersend fund, and manages local config.
Setting spending limits, auto-topup, auto-collect, and alerts still lives in the dashboard — the CLI can read those
values but not change them, and fund only prints a link to the dashboard, it does not move money.
Reference material for every flag and option is in references/commands.md. Some commands —
fund, card, marketplace, the alternate setup modes — are documented there but not in the workflows below, on
purpose: this file stays focused on the common paths. So before telling the user that ampersend can't do something, or
that a command doesn't exist, check first: run ampersend <command> --help (or bare ampersend --help), or skim
references/commands.md. Don't assert a capability is missing from memory of this file alone.
What ships with it
3 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.
- 4d ago First seen · 323 lines · 80 tokens per session scan A d750aeaf5438
ampersend is a skill published in the GitHub repository edgeandnode/ampersend-sdk (24 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 80 tokens to every session and 4,473 once invoked, about $0.0004 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 2026-08-30.
Other skills, from other repositories
compose-agents
Compose multi-agent systems with SequentialAgent, ParallelAgent, and LoopAgent. Use when building agent pipelines, parallel workflows, or iterative loops.
agent-callbacks
Add callbacks to orxhestra agents for logging, monitoring, and error handling. Covers model and tool callbacks.
agent-tools
Create and use tools with orxhestra agents. Covers functiontool, AgentTool, transfer tools, exitloop, MCP tools, and CallContext.
build-agent
Build an LLM agent with orxhestra. Use when creating a new agent, setting up LlmAgent or ReActAgent, or wiring tools to an agent.
yaml-composer
Build orxhestra agent trees from declarative YAML orx files. Covers full schema, models, tools, agents, runner, and server.
a2a-integration
Expose orxhestra agents as A2A protocol endpoints or connect to remote A2A agents.