ap2-human-not-present-flow

ap2-human-not-present-flow is a skill for Claude Code from OrcaQubits/agentic-commerce-skills-plugins. It costs 47 tokens per session (1,092 once invoked), scanned A, original, MIT.

A payment flow where a user approves shopping rules first and then leaves, allowing an agent to shop within a signed Intent Mandate. The mandate sets the limits the agent must follow.

In plain words
What is it for?
Use it to build autonomous agent purchases, enforce authorization constraints, and send unclear or unsupported cases to a merchant or user for escalation.
Why use it?
It supports shopping that continues after the user is gone while retaining a record of what the user authorized and restricting the agent's actions.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter.

Part of the ap2-agentic-payments plugin — 18 skills, 1 agent shipped together

Good fit Use it to build autonomous agent purchases, enforce authorization constraints, and send unclear or unsupported cases to a merchant or user for escalation.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/orcaqubits/agentic-commerce-skills-plugins/ap2-human-not-present-flow
Install

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.

Any agent
npx skills add OrcaQubits/agentic-commerce-skills-plugins --skill ap2-human-not-present-flow
Clone the repo
git clone --depth 1 https://github.com/OrcaQubits/agentic-commerce-skills-plugins

Made for: Claude Code.

Or install ap2-agentic-payments, the plugin that ships this one along with the rest of its 18 skills, 1 agent.

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 ap2-human-not-present-flow

README.md
[![agentmods](https://agentmods.dev/badge/skills/orcaqubits/agentic-commerce-skills-plugins/ap2-human-not-present-flow/github.svg)](https://agentmods.dev/skills/orcaqubits/agentic-commerce-skills-plugins/ap2-human-not-present-flow)
Your own site
<a href="https://agentmods.dev/skills/orcaqubits/agentic-commerce-skills-plugins/ap2-human-not-present-flow"><img src="https://agentmods.dev/badge/skills/orcaqubits/agentic-commerce-skills-plugins/ap2-human-not-present-flow/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 ap2-human-not-present-flow

Your own site · 80×15
<a href="https://agentmods.dev/skills/orcaqubits/agentic-commerce-skills-plugins/ap2-human-not-present-flow"><img src="https://agentmods.dev/badge/skills/orcaqubits/agentic-commerce-skills-plugins/ap2-human-not-present-flow.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 47 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,092 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 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 $0.00047 $0.01092
Opus 5 $0.00023 $0.00546
Sonnet 5 $0.00009 $0.00218
Haiku 4.5 $0.00005 $0.00109

Measured 9d ago against content hash 3a817f72cffa, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

ap2-human-not-present-flow 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 9d 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.

ap2-agentic-payments/skills/ap2-human-not-present-flow/SKILL.md · 103 lines

How it starts

The opening of the file, as written. The whole thing — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AP2 Human-Not-Present Transaction Flow

Before writing code

Fetch live docs:

  1. Fetch https://ap2-protocol.org/specification/ for the human-not-present flow specification
  2. Web-search site:github.com google-agentic-commerce AP2 human-not-present intent mandate for samples
  3. Fetch https://ap2-protocol.org/topics/core-concepts/ for flow overview
  4. Fetch https://ap2-protocol.org/roadmap/ and the latest spec to confirm the current support level for human-not-present (Intent Mandate) flows in the version you're targeting

Conceptual Architecture

What Human-Not-Present Means

In a human-not-present flow, the user provides authorization upfront and then leaves. The Shopping Agent acts autonomously within the constraints defined in the signed Intent Mandate.

Spec Status

Human-not-present is a first-class flow in current AP2 releases (added as the spec evolved beyond the initial human-present-only V0.1). Always verify the exact Intent Mandate fields and required signing semantics against the version you're targeting — fetch the live spec before implementing.

Flow Overview

Phase 1: Intent Capture (User Present)
  1. User → SA:       Expresses shopping intent ("Buy me running shoes under $150")
  2. SA → User:        Repeats back understanding for confirmation
  3. User:             Confirms via in-session authentication
  4. SA:               Creates Intent Mandate from user's expressed intent
  5. User → SA:        Signs the Intent Mandate (hardware-backed)
  6. User:             May go offline

Phase 2: Autonomous Shopping (User Absent)
  7. SA → Merchant:    Presents Intent Mandate
  8. Merchant:         Evaluates whether they can fulfill within constraints
  9. If uncertain →    Merchant may force user confirmation (escalate)
  10. If needs info →  Merchant asks clarification (update Intent Mandate)
  11. If can fulfill → Merchant creates Cart Mandate with offer

Phase 3: Authorization (May Require User)
  12. SA:              Evaluates Cart Mandate against Intent Mandate constraints
  13. If within bounds → SA proceeds with payment
  14. If outside bounds → SA rejects or escalates to user
  15. Payment processing (similar to human-present Phase 4-6)

Read the full file on GitHub · 103 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. 9d ago First seen · 103 lines · 47 tokens per session scan A 3a817f72cffa

Subscribe to this mod's changes

ap2-human-not-present-flow is a skill published in the GitHub repository OrcaQubits/agentic-commerce-skills-plugins (39 stars, last pushed 3mo ago), licensed MIT. It adds 47 tokens to every session and 1,092 once invoked, about $0.0002 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.

Related

Other skills, from other repositories

internet-court

Entry point for Internet Court — the trust layer for agent-to-agent commerce. Use whenever an agent needs to transact with another agent or a paid service, or a user mentions agent payments, paid APIs (HTTP 402/x402), wallet custody or trust concerns, spending mandates, delegated permissions (ERC-7710/7715), escrow…

internet-court/internet-court-skill · 147 tokens

okx-agent-payments-protocol

Use when an agent hits HTTP 402 / payment-required, or the user mentions x402, x402Version, X-PAYMENT, PAYMENT-REQUIRED, PAYMENT-SIGNATURE, WWW-Authenticate: Payment, permit2, upto, metered billing, a payment channel / voucher / session, channelId / channelid, opening / closing / topping up / settling / refunding a…

internet-court/internet-court-skill · 231 tokens

okx-dapp-discovery

Plugin router for 20 third-party DeFi protocols (Polymarket, Aave, Hyperliquid, PancakeSwap, Morpho, Raydium, Curve, Compound, Pendle, Lido, ether.fi, GMX, Kamino, Orca, Meteora, Clanker, pump.fun, Uniswap) and their protocol-native tokens (HYPE, HLP, eETH, weETH, stETH, wstETH, LDO, GHO, CAKE, CRV, COMP, RAY, ETHFI…

internet-court/internet-court-skill · 351 tokens

jupiter-lend

Interact with Jupiter Lend Protocol. Read-only SDK (@jup-ag/lend-read) for querying liquidity pools, lending markets (jlTokens), and vaults. Write SDK (@jup-ag/lend) for lending (deposit/withdraw) and vault operations (deposit collateral, borrow, repay, manage positions).

internet-court/internet-court-skill · 68 tokens

lifi

LI.FI REST API for cross-chain and same-chain token swaps, bridging, DeFi deposits (Composer), yield discovery (Earn), and intent-based execution (Intents). USE THIS SKILL WHEN USER WANTS TO: Swap tokens between different blockchains (e.g., "swap USDC on Ethereum to ETH on Arbitrum") Bridge tokens to another chain…

internet-court/internet-court-skill · 299 tokens

sales-ledger

Provides the Q3 sales ledger workbook for reporting tasks. Use when a task asks you to compute bonuses or totals over the sales ledger; the data lives only in this skill's bundled workbook.

archestra-ai/archestra · 42 tokens