Borrowing it
Nothing to install: this file belongs to mckinsey/agents-at-scale-ark. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/mckinsey/agents-at-scale-ark/main/.claude/skills/a2a-protocol/SKILL.mdgit clone --depth 1 https://github.com/mckinsey/agents-at-scale-arkWrote 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.
[](https://agentmods.dev/skills/mckinsey/agents-at-scale-ark/a2a-protocol)<a href="https://agentmods.dev/skills/mckinsey/agents-at-scale-ark/a2a-protocol"><img src="https://agentmods.dev/badge/skills/mckinsey/agents-at-scale-ark/a2a-protocol/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.
<a href="https://agentmods.dev/skills/mckinsey/agents-at-scale-ark/a2a-protocol"><img src="https://agentmods.dev/badge/skills/mckinsey/agents-at-scale-ark/a2a-protocol.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00068 | $0.02219 |
| Opus 5 | $0.00034 | $0.01110 |
| Sonnet 5 | $0.00014 | $0.00444 |
| Haiku 4.5 | $0.00007 | $0.00222 |
Grade A, and why
a2a-protocol 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 5d 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.
How it starts
The opening of the file, as written. The whole thing — 171 lines — stays where its author put it; the contents beside it link to each section on GitHub.
A2A Protocol Reference
Reference skill for the Agent2Agent (A2A) Protocol, an open standard by Google / the A2A Project (Linux Foundation) for communication between independent AI agent systems.
Attribution: All specification and topic content in
references/is sourced from the official A2A repository under the Apache-2.0 license. Each file includes source attribution in its YAML frontmatter.
IMPORTANT — Use v0.3 by default. The primary references in
references/v0.3/are v0.3.0, the version currently supported by the official Go and Python SDKs and most implementations. The v1.0 RC spec is inreferences/v1.0-rc/and should only be consulted when explicitly working on v1.0 features or migration planning.
When to use this skill
- Building or debugging an A2A server or client
- Configuring Ark
A2AServercustom resources - Understanding Agent Cards, task lifecycle, or message formats
- Implementing streaming (SSE) or push notifications
- Working with A2A extensions
- Comparing A2A with MCP
Protocol overview
A2A enables agents built on different frameworks to discover capabilities, negotiate interaction modes, manage collaborative tasks, and exchange information — without exposing internal state, memory, or tools.
Core actors:
- User — human or automated service initiating a request
- A2A Client — application or agent acting on behalf of the user
- A2A Server — agent exposing an HTTP endpoint implementing A2A
Core elements:
| Element | Purpose |
|---|---|
| Agent Card | JSON metadata: identity, capabilities, endpoint, auth |
| Task | Stateful unit of work with unique ID and lifecycle |
| Message | Single communication turn (role: "user" or "agent") |
| Part | Content container: text, file reference, or structured data |
| Artifact | Tangible output generated during a task |
What ships with it
45 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/v0.3/spec-01-introduction.md 2.6 KB
- references/v0.3/spec-02-core-concepts.md 1.9 KB
- references/v0.3/spec-03-transport.md 10 KB
- references/v0.3/spec-04-authentication.md 4.4 KB
- references/v0.3/spec-05-agent-card.md 10.0 KB
- references/v0.3/spec-06-data-objects.md 6.4 KB
- references/v0.3/spec-07-rpc-methods.md 17 KB
- references/v0.3/spec-08-error-handling.md 5.2 KB
- references/v0.3/spec-09-workflows.md 20 KB
- references/v0.3/spec-10-appendices.md 3.4 KB
- references/v0.3/spec-11-compliance.md 5.2 KB
- references/v0.3/topic-a2a-and-mcp.md 7.2 KB
- references/v0.3/topic-agent-discovery.md 7.4 KB
- references/v0.3/topic-enterprise-ready.md 8.4 KB
- references/v0.3/topic-extensions.md 16 KB
- references/v0.3/topic-key-concepts.md 7.5 KB
- references/v0.3/topic-life-of-a-task.md 9.5 KB
- references/v0.3/topic-streaming-and-async.md 15 KB
- references/v0.3/topic-what-is-a2a.md 5.8 KB
- references/v1.0-rc/spec-01-introduction.md 6.5 KB
- references/v1.0-rc/spec-02-terminology.md 2.2 KB
- references/v1.0-rc/spec-03-operations.md 39 KB
- references/v1.0-rc/spec-04-data-model.md 11 KB
- references/v1.0-rc/spec-05-binding-requirements.md 10 KB
- references/v1.0-rc/spec-06-workflows.md 12 KB
- references/v1.0-rc/spec-07-authentication.md 2.6 KB
- references/v1.0-rc/spec-08-agent-card.md 11 KB
- references/v1.0-rc/spec-09-jsonrpc-binding.md 6.9 KB
- references/v1.0-rc/spec-10-grpc-binding.md 6.9 KB
- references/v1.0-rc/spec-11-http-rest-binding.md 7.2 KB
- references/v1.0-rc/spec-12-custom-binding.md 5.0 KB
- references/v1.0-rc/spec-13-security.md 9.4 KB
- references/v1.0-rc/spec-14-iana.md 5.2 KB
- references/v1.0-rc/spec-appendix-a-migration.md 10 KB
- references/v1.0-rc/spec-appendix-b-mcp.md 1.6 KB
- references/v1.0-rc/topic-a2a-and-mcp.md 7.0 KB
- references/v1.0-rc/topic-agent-discovery.md 6.0 KB
- references/v1.0-rc/topic-definitions.md 1.0 KB
- references/v1.0-rc/topic-enterprise-ready.md 7.9 KB
- references/v1.0-rc/topic-extensions.md 13 KB
- references/v1.0-rc/topic-key-concepts.md 6.8 KB
- references/v1.0-rc/topic-life-of-a-task.md 11 KB
- references/v1.0-rc/topic-streaming-and-async.md 10 KB
- references/v1.0-rc/topic-what-is-a2a.md 11 KB
- references/v1.0-rc/topic-whats-new-v1.md 28 KB
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.
- 5d ago First seen · 171 lines · 68 tokens per session scan A 2f6334f3e558
a2a-protocol is a skill published in the GitHub repository mckinsey/agents-at-scale-ark (422 stars, last pushed today), licensed Apache-2.0. It adds 68 tokens to every session and 2,219 once invoked, about $0.0003 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 2026-09-03.
Other skills, from other repositories
container-manager-kubernetes-operations
Full operational Kubernetes surface via the container-manager-mcp MCP server — workloads (pods/rollouts/StatefulSets/DaemonSets/ReplicaSets/Jobs/CronJobs), config (ConfigMaps/Secrets/Namespaces/CRDs/patch), networking (Ingress/native Services/NetworkPolicy/DNS), storage (PV/PVC/StorageClass/snapshots/CSI), RBAC…
k8s-ops
Opinionated multi-step Kubernetes workflows on top of the k8s-mcp server. Encodes tool-sequencing logic and failure-mode decision trees for deploying from a repo, debugging pods/rollouts, performing safe restarts, and auditing cluster posture. Use when the user asks to deploy, diagnose, restart, roll out, or audit…
kubernetes-mesh-provisioner
Kubernetes Mesh Provisioner atomic skill. Stands up an RKE2 cluster (server + agents) with Cilium CNI and the NVIDIA GPU device plugin, the Kubernetes parallel of swarm-mesh-provisioner. Idempotent — re-runnable.
cost-optimization
Optimize cloud costs across AWS, Azure, GCP, and OCI through resource rightsizing, tagging strategies, reserved instances, and spending analysis. Use when reducing cloud expenses, analyzing infrastructure costs, or implementing cost governance policies.
network-rca
Kubernetes network root cause analysis skill powered by Kubeshark MCP. Use this skill whenever the user wants to investigate past incidents, perform retrospective traffic analysis, take or manage traffic snapshots, extract PCAPs, dissect L7 API calls from historical captures, compare traffic patterns over time, detect…
linkerd-patterns
Implement Linkerd service mesh patterns for lightweight, security-focused service mesh deployments. Use when setting up Linkerd, configuring traffic policies, or implementing zero-trust networking with minimal overhead.