A Rust toolkit and workflow for building MCP servers and clients. MCP, or Model Context Protocol, is a standard way for AI applications to expose and use tools, prompts, and data.

Creating MCP servers and clients in Rust, exposing tools and resources, handling requests, configuring transports and authentication, and deploying MCP services.

It helps developers use the correct protocol version and avoid writing code for an older, incompatible interface.

How ferrus-dev/ferrus configures Claude Code and Codex

Reuse

Borrowing it

Nothing to install: this file belongs to ferrus-dev/ferrus. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/ferrus-dev/ferrus/main/.agents/skills/neva/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/ferrus-dev/ferrus

Made for: Claude Code, Codex.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/ferrus-dev/ferrus/neva/github.svg)](https://agentmods.dev/skills/ferrus-dev/ferrus/neva)

Your own site

<a href="https://agentmods.dev/skills/ferrus-dev/ferrus/neva"><img src="https://agentmods.dev/badge/skills/ferrus-dev/ferrus/neva/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/ferrus-dev/ferrus/neva"><img src="https://agentmods.dev/badge/skills/ferrus-dev/ferrus/neva.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 139 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 4,064 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00139	$0.04064
Opus 5.5	$0.00056	$0.01626
Sonnet 5.5	$0.00028	$0.00813
Haiku 4.5	$0.00014	$0.00406

Measured 21d ago against content hash c89076704a4d, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

neva 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 21d 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.

.agents/skills/neva/SKILL.md · 314 lines

How it starts

The opening of the file, as written. The whole thing — 314 lines — stays where its author put it; the contents beside it link to each section on GitHub.

neva — MCP servers and clients in Rust

neva is a Rust SDK for the Model Context Protocol. One crate covers both sides: App builds servers, Client builds clients, and a single process can run both.

This skill describes neva 0.8.x, which speaks MCP 2026-07-28 by default. That protocol generation broke compatibility with everything before it. Most MCP knowledge in circulation — and every other MCP SDK — describes the older generation, so the failure mode here is not "you forget an API", it is "you confidently write the previous protocol". The Non-negotiables section below is the list of places where that happens. Read it before writing code, every time.

Step 1 — establish the version and the profile

Do this first; it changes which API is correct.

cargo add neva --features server-full   # server
cargo add neva --features client-full   # client
cargo add neva --features full          # both

In an existing project, read Cargo.toml:

What you find	What it means
`neva = "0.8"` and no `legacy-spec`	Default profile, MCP 2026-07-28. This skill applies as written
`neva = "0.7"` and no `legacy-spec`	As 0.8, minus knowing the caller: no `ctx.claims()`, no `scopes = [..]` gates, no `with_decoding_key` / `with_jwks_uri`. `DefaultClaims.aud` is an `Option<String>`, `with_alg` adds to `HS256` instead of replacing it, and a middleware's answer without `next` never reaches the caller. `references/legacy.md` has the upgrade
`neva = "0.6"` and no `legacy-spec`	Same protocol, but the client and `Context` calls are flat — `client.call_tool(..)`, `ctx.find_tool(..)` — where 0.7+ has namespaces. Write the flat ones on 0.6; no `svir` feature. `registry` and a retryable `Client::connect` need 0.6.1, not 0.6.0
`neva = "0.5"` and no `legacy-spec`	As 0.6, minus synchronous handlers and per-request extensions; see Handler shapes for what 0.6 added
`features = [… "legacy-spec" …]`	Legacy profile, MCP 2024-11-05 … 2025-11-25. A different API. Read `references/legacy.md` before touching anything
`neva = "0.4"` or older	Pre-2026-07-28 by default. Read `references/legacy.md` for the upgrade
`proto-2026-07-28-rc`	A flag that no longer exists — remove it

Read the full file on GitHub · 314 lines

Files

What ships with it

9 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.

Changes

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.

21d ago Changed · +28 lines · +25 tokens per session c89076704a4d
1mo ago Changed · +75 lines · +20 tokens per session 0817ef1cd97a
2mo ago First seen · 211 lines · 94 tokens per session scan A 11ea16a79be0

Subscribe to this mod's changes

neva is a skill published in the GitHub repository ferrus-dev/ferrus (2 stars, last pushed 7d ago), licensed Apache-2.0. It adds 139 tokens to every session and 4,064 once invoked, about $0.0006 per session on Opus 5.5. 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 2026-09-04.

Building agents Languages Rust agent-orchestration ai-agents automation deterministic-systems developer-tools mcp

neva next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
nevaferrus-dev/ferrus	Skill	2	7d ago	Apache-2.0	A	139	Claude Code, Codex
rust-checkHmbown/CodeWhale	Skill	41k	5d ago	MIT	A	15	Claude Code, Codex
Rust Standardssystempromptio/systemprompt-template	Skill	29	9d ago	none	A	23	Claude Code
agentic-infrastructure-2026curiositech/windags-skills	Skill	18	6d ago	none	A	173	Claude Code
rust-safetyLu1sDV/skillsmd	Skill	2	today	none	A	125	Claude Code, Codex
rust-performanceLu1sDV/skillsmd	Skill	2	today	none	A	91	Claude Code, Codex
rust-algorithmsLu1sDV/skillsmd	Skill	2	today	none	A	94	Claude Code, Codex

ferrus: Skill for Claude Code