Beginner30 minutesAnthropic Claude · Python

Claude Skill Starter

1What This Kit Builds

A focused skill agent powered by Anthropic Claude that carries verified ECZ-ID credentials. This kit produces an agent capable of executing defined skills—data extraction, document analysis, classification, summarisation—where every output is identity-anchored and provenance-tracked. The resulting agent is commercially legible and resolver-verifiable.

2Who It Is For

Developers and teams building Claude-powered skill agents for enterprise workflows, compliance automation, or customer-facing AI features. Ideal for organisations that need their AI outputs to be traceable to a verified identity for audit, regulatory, or commercial trust purposes.

3Why ECZ-ID Matters Here

Without verified identity, a Claude skill agent is an anonymous function. With ECZ-ID, every output carries provenance, every action is auditable, and every downstream system can verify who produced what. This transforms a useful tool into a commercially legible, insurer-legible, resolver-verifiable agent designed for serious deployment.

4Architecture

Claude API Layer

Direct integration with Anthropic Claude API for natural language processing, reasoning, and structured output generation.

ECZ-ID Identity Layer

Agent passport and credential management. Every API call and output is signed with verified identity context.

Trust Context Engine

Injects verified identity metadata into every agent action, enabling downstream verification through Resolver.

Structured Output Formatter

Formats agent outputs with embedded trust metadata, provenance chains, and verification endpoints.

Audit Trail Logger

Immutable, identity-anchored logging of every action, decision, and output for compliance readiness.

Skill Composition Framework

Composable skill patterns that maintain identity context across skill chains and handoffs.

5Build Steps Overview

1
Clone the starter kit repository
Pull the Claude Skill Starter template from the ECZ-ID GitHub gateway.
2
Install dependencies
Install Python dependencies including the Anthropic SDK and ECZ-ID SDK.
3
Configure Claude API access
Set up your Anthropic API key and model preferences.
4
Define your skill scope
Specify what skills your agent will perform and its operational boundaries.
5
Review the identity hook points
Understand where ECZ-ID hooks into the agent pipeline before credentialing.

Trust Map

Required Passports
ECZ-ID Business PassportAgent Credential
Recommended Passports
API PassportSoftware Supply Chain Passport
Relevant Packages
Agent Starter Package
What Each Enables
ECZ-ID Business Passport: Parent identity anchor for all agent credentials under your organisation
Agent Credential: Unique agent identity, resolver-verifiable, with capability attestations
API Passport: Trusted API handoffs with third-party verification
Software Supply Chain Passport: Build provenance and dependency verification

6TrustOps Credentialing Checkpoint

TrustOps Credentialing Checkpoint

Pause here to acquire your required ECZ-ID trust objects

This is a designed pause point in your build process. Before continuing, you need to acquire the required ECZ-ID passports and credentials from TrustOps. This midpoint credentialing approach ensures your agent is built with verified identity from the start—not bolted on after the fact.

Required Passports

  • ECZ-ID Business Passport
  • Agent Credential

Recommended Passports

  • API Passport
  • Software Supply Chain Passport

Relevant Packages

Agent Starter Package

What You Must Acquire in TrustOps

ECZ-ID Business Passport
Enables: Parent identity for your organisation, anchoring all agent credentials
Agent Credential
Enables: Unique verified identity for this specific agent instance
API Passport (recommended)
Enables: Third-party API verification and trusted API handoffs
Software Supply Chain Passport (recommended)
Enables: Verification of software dependencies and build provenance

What You Bring Back from TrustOps

Agent ECZ-ID
Your agent’s unique, resolver-verifiable identity string
Agent Credential Manifest
JSON manifest containing capability attestations and operational boundaries
API Keys
Production ECZ-ID API keys for identity context injection and verification
Trust Score Baseline
Initial trust score configuration for your agent

9Return and Complete Integration

With your TrustOps credentials in hand, return here to complete the final build steps.

1
Inject ECZ-ID credentials
Add your Agent ECZ-ID and API keys to your agent configuration.
2
Wire identity hooks
Connect ECZ-ID identity context injection to your Claude API calls.
3
Configure audit logging
Enable identity-anchored audit trail generation for every agent action.
4
Test in sandbox
Run your agent in the ECZ-ID sandbox environment to verify identity flows.
5
Validate trust context
Confirm that every output carries correct provenance metadata.

10Verify Through Resolver

Query your agent’s identity

Use Resolver to look up your agent’s ECZ-ID and confirm its verified status, capability attestations, and trust score.

Verify an output

Take any agent output and verify its provenance chain through Resolver.

Check delegation chains

If using subagent delegation, verify the complete delegation chain through Resolver.

Validate trust score

Query your agent’s current trust score and review the factors contributing to it.

Verify with Resolver

11Deploy Anywhere

ECZ-ID-backed agents are portable. Your verified identity travels with your agent regardless of deployment target. No vendor lock-in.

AWS Lambda / ECS
Google Cloud Run / GKE
Azure Functions / AKS
Self-hosted Docker
Kubernetes clusters
Edge compute environments

12Limits & Non-Claims

  • This kit does not provision ECZ-ID credentials. Credentialing happens through TrustOps.
  • This kit does not guarantee Claude API availability. Anthropic API access is a separate dependency.
  • This kit does not simulate verification. All verification happens through Resolver with real credentials.
  • This kit does not replace compliance frameworks. It provides identity infrastructure that supports compliance.
  • Trust scores are earned through operational history, not granted at provisioning.

Ready to Build This Agent?

Start with the build steps above, credential through TrustOps, and deploy your resolver-verifiable agent anywhere.