# Agent To Agent Auth

> Enforce SPIFFE/SPIRE plus mTLS for agent-to-agent communication. Use when designing, building, or reviewing authentication between AI agents or multi-agent systems.

- **Type:** Skill
- **Install:** `agentstack add skill-redhatproductsecurity-prodsec-skills-agent-to-agent-auth`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [RedHatProductSecurity](https://agentstack.voostack.com/s/redhatproductsecurity)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** Apache-2.0
- **Upstream author:** [RedHatProductSecurity](https://github.com/RedHatProductSecurity)
- **Source:** https://github.com/RedHatProductSecurity/prodsec-skills/tree/main/module/skills/agent-to-agent-auth

## Install

```sh
agentstack add skill-redhatproductsecurity-prodsec-skills-agent-to-agent-auth
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Agent-to-Agent Authentication

## Security Requirement

Agent-to-agent communication MUST follow the service-to-service recommendation and implement SPIFFE/SPIRE+mTLS.

## Why SPIFFE/SPIRE for Agent-to-Agent

- Agents are software workloads, making SPIFFE the natural identity framework
- No credentials exchanged over the network (SPIRE handles attestation and certificate issuance)
- Mutual authentication ensures both agents verify each other's identity
- Short-lived X.509 certificates (SVIDs) are automatically rotated
- Each agent gets a unique SPIFFE ID that can be used for authorization decisions

## Architecture

```
Agent A                          Agent B
  │                                │
  ├── SPIFFE ID:                   ├── SPIFFE ID:
  │   spiffe://domain/agent/a      │   spiffe://domain/agent/b
  │                                │
  ├── Gets SVID from SPIRE Agent   ├── Gets SVID from SPIRE Agent
  │                                │
  └── mTLS connection ────────────→└── Validates Agent A's SVID
      (presents SVID)                  (presents own SVID)
```

## SPIFFE ID Convention for Agents

Use a consistent SPIFFE ID naming convention:

```
spiffe:///agent//
```

Examples:
- `spiffe://example.com/agent/data-analyst/prod-01`
- `spiffe://example.com/agent/code-reviewer/staging-02`

## Implementation Checklist

- [ ] Deploy SPIRE server and agents for the agent infrastructure
- [ ] Register each agent type as a SPIRE workload entry
- [ ] Configure agents to obtain SVIDs from the SPIRE agent
- [ ] Implement mTLS for all agent-to-agent communication
- [ ] Extract peer agent's SPIFFE ID from the TLS certificate for authorization
- [ ] Define authorization policies based on SPIFFE IDs (which agents can communicate)
- [ ] Monitor SVID rotation and certificate expiration

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [RedHatProductSecurity](https://github.com/RedHatProductSecurity)
- **Source:** [RedHatProductSecurity/prodsec-skills](https://github.com/RedHatProductSecurity/prodsec-skills)
- **License:** Apache-2.0

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-redhatproductsecurity-prodsec-skills-agent-to-agent-auth
- Seller: https://agentstack.voostack.com/s/redhatproductsecurity
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.