schema-designer

schema-designer is an agent for Claude Code from kitchen-engineer42/joharnessburg. It costs 88 tokens per session (873 once invoked), scanned A, original, MIT.

A schema-design agent for deciding how information should be structured by studying several representative source samples, including unusual cases.

In plain words
What is it for?
Use it during knowledge-schema planning to propose, test, refine, and record the format an extractor will use for facts, rules, slide concepts, wiki content, or mixed material.
Why use it?
It helps avoid a structure that works for one example but fails when applied consistently across the rest of a collection.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter; mentions subagents.

Part of the john plugin — 28 skills, 5 commands, 5 agents, 3 hooks shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/kitchen-engineer42/joharnessburg/schema-designer
Clone the repo
git clone --depth 1 https://github.com/kitchen-engineer42/joharnessburg

Made for: Claude Code.

Or install john, the plugin that ships this one along with the rest of its 28 skills, 5 commands, 5 agents, 3 hooks.

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 schema-designer

README.md
[![agentmods](https://agentmods.dev/badge/agents/kitchen-engineer42/joharnessburg/schema-designer.svg)](https://agentmods.dev/agents/kitchen-engineer42/joharnessburg/schema-designer)
Your own site
<a href="https://agentmods.dev/agents/kitchen-engineer42/joharnessburg/schema-designer"><img src="https://agentmods.dev/badge/agents/kitchen-engineer42/joharnessburg/schema-designer.svg" alt="Measured on agentmods" height="20"></a>
Per session 88 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 873 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00088 $0.00873
Opus 5 $0.00044 $0.00436
Sonnet 5 $0.00018 $0.00175
Haiku 4.5 $0.00009 $0.00087

Measured 6d ago against content hash c32529f196ca, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

schema-designer 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 6d 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.

plugins/joharnessburg/agents/schema-designer.md · 47 lines

How it starts

The opening of the file, as written. The whole thing — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.

schema-designer

You are dispatched when the app-type definition cascade (knowledge format → knowledge schema → app mechanism → build pipeline) needs deliberate schema work — when reading one chunk and guessing won't produce a schema the extractor can apply consistently across the rest of the corpus.

What you receive in your prompt

  • The project's knowledge format: facts / rules / slide-concepts / wiki / mixed. This is settled upstream; don't re-litigate.
  • A representative sample of source chunks: 3–10 chunks the user (or upstream phase) has flagged as covering the diversity of the source — favor the corpus's edge cases over its average (the weird chapter breaks schemas; the typical one proves nothing). Paths or excerpts.
  • The project intent from PLAN.md's top: what the produced app does, who uses it, what success looks like. Schema must serve this.
  • Any template constraints: e.g., a doc-verification project may lock the schema to rules + glossary; in that case your job is to confirm or surface incompatibility, not propose alternatives.
  • The output target: a markdown file path or a section of PLAN.md to write the settled schema into.

What you produce

A schema proposal in the format the project's schema-design skill expects. Typically:

  1. Field list with type, required/optional, one-line purpose per field.
  2. Header vs body split (progressive disclosure — what shows in lists / search results vs full entry).
  3. MECE check: a paragraph explaining how the schema avoids ambiguity (two entries can't both describe the same precondition+verdict) and covers the source (the sample chunks don't have content that escapes the schema).
  4. Open questions for the user: any decisions you couldn't make autonomously.

Iteration discipline

You may use up to 3 internal turns to refine — read sample chunks, draft, mentally apply the draft to other chunks, revise. After 3 turns, return what you have plus the open questions. Don't loop forever; the user is the tiebreaker.

Read the full file on GitHub · 47 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. 6d ago First seen · 47 lines · 88 tokens per session scan A c32529f196ca

Subscribe to this mod's changes

schema-designer is an agent published in the GitHub repository kitchen-engineer42/joharnessburg (9 stars, last pushed 1mo ago), licensed MIT. It adds 88 tokens to every session and 873 once invoked, about $0.0004 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 2026-08-31.

Related

Other agents, from other repositories

ai-agent

AI feature implementation specialist. Handles STT, LLM, and AI service integration with context-aware patterns. Auto-discovers project conventions before implementing. Supports OpenAI, Anthropic, and other AI providers with streaming, error handling, and cost optimization.

wigtn/wigtn-plugins · 54 tokens

unfolding-po

PO (Product Owner) role in the Unfolding Specs process. Decomposes Features into smaller Features, creates Acceptance Tests, and identifies implicit business assumptions as Domain Model Decisions (DMDs).

t1/tdder · 45 tokens

mlops

ML infrastructure specialist — training pipelines, model serving, GPU optimization, distributed training.

cdeust/zetetic-team-subagents · 18 tokens

experiment-runner

ML/empirical experiment design specialist — pre-registration, Fisher-style design, reproducibility manifests.

cdeust/zetetic-team-subagents · 22 tokens

prompt-engineer

Use when: creating new prompts, optimizing existing prompts, reviewing prompt quality, designing agents or skills. Do NOT use for: code implementation (use domain expert), non-prompt tasks.

fusengine/agents · 41 tokens

evolve-data-integrity-check

Data-pipeline integrity auditor for the Evolve Loop (Evaluate archetype). The advisor INSERTS this phase after Build whenever the cycle's scout.goaltype == "data-pipeline", to statically audit the changed batch/stream code for records it could silently corrupt, drop, duplicate, or reorder — and BLOCKS when a CRITICAL…

mickeyyaya/evolve-loop · 83 tokens