From ap2-agentic-payments
Implements AP2 risk signals framework for agentic payments, covering novel risks like delegated trust and user asynchronicity, risk payload construction, and fraud assessment.
npx claudepluginhub orcaqubits/agentic-commerce-skills-plugins --plugin ap2-agentic-paymentsThis skill is limited to using the following tools:
**Fetch live docs**:
Guides Next.js Cache Components and Partial Prerendering (PPR) with cacheComponents enabled. Implements 'use cache', cacheLife(), cacheTag(), revalidateTag(), static/dynamic optimization, and cache debugging.
Migrates code, prompts, and API calls from Claude Sonnet 4.0/4.5 or Opus 4.1 to Opus 4.5, updating model strings on Anthropic, AWS, GCP, Azure platforms.
Automates semantic versioning and release workflow for Claude Code plugins: bumps versions in package.json, marketplace.json, plugin.json; verifies builds; creates git tags, GitHub releases, changelogs.
Fetch live docs:
https://ap2-protocol.org/specification/ for risk payload specificationhttps://ap2-protocol.org/topics/privacy-and-security/ for risk considerationsap2 protocol risk signals fraud agentic payments for risk framework detailssite:github.com google-agentic-commerce AP2 risk for implementation referencesAgentic commerce introduces novel risk dimensions that traditional payment systems weren't designed for. AP2's risk signals framework provides a common language for all ecosystem participants to assess transaction risk.
| Risk Factor | Description |
|---|---|
| User asynchronicity | User may not be present during the entire transaction journey |
| Delegated trust | Agents initiate transactions on behalf of users |
| Mandate-merchant matching | Verifying the purchase matches the authorized intent |
| Temporal gaps | Time between token generation and payment execution |
| Indirect trust establishment | CP and Merchant may not have a direct trust relationship |
| Agent identity verification | Verifying the agent is who it claims to be |
The following are additional AI-specific risk considerations relevant to agentic commerce implementations, but they are not part of the official AP2 specification's novel risk factor table:
| Risk Factor | Description |
|---|---|
| Agent hallucination | AI agent may misinterpret user intent |
| Prompt injection | Malicious inputs that manipulate agent behavior |
The risk payload is an open-ended field structure in V0.1:
AP2 defines trust establishment phases:
Short-term (V0.1):
Long-term (future):
For dispute resolution, risk signals help determine accountability:
Fetch the specification for exact risk payload structure, supported signal types, and risk assessment requirements before implementing.