16 specifications published

Agent Standards

No standards body has published AI agent protocols. We're proposing the first — open, practical, and submission-ready.

The Gap

Every major standards body has published AI guidelines. None address how agents talk to each other, verify identity, transfer skills, or build trust.

ISO/IEC

18 AI standards — data quality, lifecycle, risk. Zero about agent-to-agent protocols.

CEN-CENELEC

AI Act harmonised standards in development. All about human safety. No agent compliance automation.

IEEE

7000-series for ethical AI. Nothing for agent skill formats or cross-platform interoperability.

IETF

Zero AI agent RFCs. MCP is not an IETF standard. The agent communication layer is completely unbuilt.

NIST

AI Risk Management Framework exists. No agent-specific profile. No agent identity verification.

BSI

AI ethics PAS published. No agent identity or trust framework. UK-first opportunity unclaimed.

Every body is focused on AI-for-humans. Nobody is doing AI-for-AI. That's the territory we're claiming.

Priority Submissions

Six standards bodies, six specifications. Ordered by impact and barrier to submission.

🥇
IETF
IACP — Inter-Agent Communication Protocol
Internet-Draft for agent-to-agent messaging. Every agent that talks over the web uses it. Lowest barrier — individuals can submit Internet-Drafts directly.
🥇
CEN-CENELEC
Compliance-as-Code
CEN Workshop Agreement for automated AI Act compliance verification. Mandatory for every AI agent deployed in the EU. First-mover advantage in a regulatory gap.
🥈
BSI
Agent Identity Protocol
PAS for cryptographically verifiable agent identity. UK-first standard — then push to ISO globally. No agent identity standard exists anywhere.
🥈
IEEE
ASFS — Agent Skill Format Standard
Cross-platform skill portability. Write a skill once, run it on any agent platform. IEEE's global reach makes this the default for enterprise procurement.
🥉
ISO
Agent Handoff Protocol
New Work Item Proposal for standardised agent task transfer. 3–5 year global standard. The long game — ISO adoption means mandatory compliance in government procurement worldwide.
🥉
NIST
AI RMF Agent Profile
Agent-specific extension to the AI Risk Management Framework. The path to US federal government adoption. Every gov contractor running agents will need this.

All 16 Specifications

Full specs at workswithagents.dev/specs/. All CC BY 4.0 — free to use, adapt, and submit to standards bodies.

🗣 IACP

Inter-Agent Communication Protocol. Standardised messaging, discovery, and handshake between agents across platforms.

📦 ASFS

Agent Skill Format Standard. Write a skill once — run it on any agent platform. Frontmatter spec, validation rules, converter tooling.

✅ Compliance-as-Code

Automated EU AI Act, NHS DTAC, GDPR, and FCA compliance verification. Machine-readable regulation packs.

🆔 Agent Identity Protocol

Cryptographically verifiable agent identity. DID-based. Cross-platform authentication without central authority.

🤝 Agent Handoff Protocol

Standardised task transfer between agents. State serialisation, capability matching, audit trails.

💰 Agent Economics Protocol

Agent-to-agent service discovery, pricing, and payment. Token metering, SLA enforcement, marketplace mechanics.

⭐ Reputation Ledger

Immutable agent performance history. Peer-reviewed outcomes, skill endorsements, trust accrual over time.

🛡 Trust Score Framework

Multi-factor agent trust scoring. Identity strength, performance history, compliance status, peer endorsements.

🏗 Fleet Insurance

Observability, resilience, and insurance mechanics for multi-agent deployments. Health monitoring, failover, coverage tiers.

📡 Agent OSI Model

7-layer model for agent infrastructure — from physical compute through application. Maps to existing networking standards.

📋 Capability Manifest

Machine-readable declaration of agent capabilities. Tools, models, skills, rate limits, safety constraints.

🔗 Coordination Protocol

Multi-agent task decomposition and coordination. Parallel workstreams, dependency resolution, merge strategies.

⚖ SLA Framework

Standardised agent service level agreements. Uptime, latency, accuracy, and recourse — machine-readable and enforceable.

🚀 Deployment Manifest

One-file agent deployment specification. Hardware requirements, model configs, network needs, security policies.

👋 Onboarding Protocol

Standardised agent registration across platforms. Identity bootstrap, capability announcement, credential exchange.

🔁 Transaction Protocol

Atomic, auditable agent-to-agent transactions. Request, execution, verification, settlement — with rollback.

Why This Matters

🔓 No Vendor Lock-In

Standardised skills and communication mean agents work across platforms. Switch providers without rebuilding everything.

📋 Auditable Compliance

Machine-readable regulation packs mean every agent action is verifiable against GDPR, NHS DTAC, EU AI Act, FCA rules.

🆔 Verifiable Identity

Cryptographic agent identity means you know which agent did what — essential for regulated industries and enterprise procurement.

🔄 Portable Skills

A skill built for one agent platform should work on any other. ASFS makes agent investments durable across the ecosystem.

🤝 Agent Economy

Without standards, agents can't discover, pay, or rate each other. The agent-to-agent economy needs protocols to function.

🏛 Government-Ready

NHS, MOD, HMRC procurement requires standards. Without them, AI agents stay locked out of the public sector.

These are proposals, not pronouncements

We're submitting them to the bodies that matter. If you're building agent infrastructure, read the specs and get involved.

Read the Specs →

All specifications CC BY 4.0. Pull requests welcome. Certification criteria →