kimojio-rs: Skill for Claude Code

.github/skills/kimojio-json-decoder/SKILL.md

kimojio-json-decoder is a skill for Claude Code, Codex from Azure/kimojio-rs. It costs 70 tokens per session (2,468 once invoked), scanned A, original, MIT.

A guide for writing Rust code that turns tokenized JSON into typed values without allocating heap memory. JSON is a common text format for structured data, and a tokenizer breaks it into readable pieces.

In plain words
What is it for?
Use it to build or change decoders for structs, enums, or iterators, including rules for optional or unknown fields, nesting limits, offsets, and completion.
Why use it?
It helps preserve input lifetimes, memory limits, error details, and interface requirements while avoiding hand-written state-machine mistakes.

Skill for Claude CodeCodex ✓ vendor

Written for no agent in particular: nothing here depends on one.

This is Azure/kimojio-rs's own configuration. It tells Claude Code and Codex how to work on kimojio-rs itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything kimojio-rs configures →

About the project

Kimojio is a Linux asynchronous runtime for Rust that uses one core per thread and io_uring for disk input and output, with cooperative task scheduling. It suits I/O-bound programs that need predictable scheduling and gives developers explicit control over concurrency and load distribution.

Azure/kimojio-rs · 262 stars · on GitHub

Reuse

Borrowing it

Nothing to install: this file belongs to Azure/kimojio-rs. 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/Azure/kimojio-rs/main/.github/skills/kimojio-json-decoder/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/Azure/kimojio-rs

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.

agentmods badge for kimojio-json-decoder

README.md
[![agentmods](https://agentmods.dev/badge/skills/azure/kimojio-rs/kimojio-json-decoder/github.svg)](https://agentmods.dev/skills/azure/kimojio-rs/kimojio-json-decoder)
Your own site
<a href="https://agentmods.dev/skills/azure/kimojio-rs/kimojio-json-decoder"><img src="https://agentmods.dev/badge/skills/azure/kimojio-rs/kimojio-json-decoder/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.

agentmods 80×15 button for kimojio-json-decoder

Your own site · 80×15
<a href="https://agentmods.dev/skills/azure/kimojio-rs/kimojio-json-decoder"><img src="https://agentmods.dev/badge/skills/azure/kimojio-rs/kimojio-json-decoder.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 70 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,468 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. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • medium Excessive Agency · line 38
    Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
    Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
How audits are shown
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00070 $0.02468
Opus 5 $0.00035 $0.01234
Sonnet 5 $0.00014 $0.00494
Haiku 4.5 $0.00007 $0.00247

Measured 4d ago against content hash 4c6a44bd267b, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

kimojio-json-decoder 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 4d 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.

.github/skills/kimojio-json-decoder/SKILL.md · 253 lines

How it starts

The opening of the file, as written. The whole thing — 253 lines — stays where its author put it; the contents beside it link to each section on GitHub.

kimojio-json decoders

kimojio-json tokenizes JSON. It reports borrowed input to a visitor. A decoder maps those events to a typed Rust value.

This skill is for direct, allocation-free decoders. It gives the agent the constraints and the reasons for them. It does not prescribe one state layout for every schema. A previous implementation is evidence, not a specification. If a task gives a type or an interface, treat that interface as a contract unless the task also asks to change it.

Start with the problem

Before writing code, identify:

  • the input source and schema;
  • the required public interface and input lifetime;
  • whether the result is one value, a sequence, or a callback stream;
  • the policy for unknown members and optional members;
  • maximum sizes and nesting depth;
  • error kinds, byte offsets, and completion behavior;
  • the expected performance and trust boundary.

Then compare possible designs. For each design, ask:

  1. What work or failure mode does it avoid?
  2. Does that cost apply to this schema and interface?
  3. What is the smallest representation that satisfies the contract?
  4. What measurement or test would distinguish the choices?

State the reason for a choice in the module documentation or change summary. Do not copy a detailed technique from an earlier decoder without checking why it was used. A different schema may make another technique better.

Read the tokenizer

Read these files from the repository root before relying on an API detail:

  • kimojio-json/src/lib.rs
  • kimojio-json/src/tokenizer.rs
  • kimojio-json/src/escape_string.rs
  • kimojio-json/src/error.rs

Confirm names, lifetimes, return values, and error behavior in the source. Do not invent a pull-token API or assume that a callback has access to the tokenizer.

Prefer the tokenizer's visitor protocol when it fits

The tokenizer already owns the scan loop and calls a Visitor. Prefer this protocol when the decoder can consume events as they arrive. This avoids an extra token iterator and an extra dispatch layer.

Read the full file on GitHub · 253 lines

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.

  1. 4d ago First seen · 253 lines · 70 tokens per session scan A 4c6a44bd267b

Subscribe to this mod's changes

kimojio-json-decoder is a skill published in the GitHub repository Azure/kimojio-rs (262 stars, last pushed 28d ago), licensed MIT. It adds 70 tokens to every session and 2,468 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 2026-09-04.