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.
npx agentmods add instructions/kaptinlin/jsonschema/claude-mdgit clone --depth 1 https://github.com/kaptinlin/jsonschemaWrote 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/instructions/kaptinlin/jsonschema/claude-md)<a href="https://agentmods.dev/instructions/kaptinlin/jsonschema/claude-md"><img src="https://agentmods.dev/badge/instructions/kaptinlin/jsonschema/claude-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.02208 | $0.02208 |
| Opus 5 | $0.01104 | $0.01104 |
| Sonnet 5 | $0.00442 | $0.00442 |
| Haiku 4.5 | $0.00221 | $0.00221 |
Grade A, and why
jsonschema CLAUDE.md 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 — 180 lines — stays where its author put it; the contents beside it link to each section on GitHub.
jsonschema
High-performance JSON Schema Draft 2020-12 validation for Go. The package combines a compiler, a direct validation API for JSON/maps/structs, default-aware unmarshaling, a fluent constructor API, and struct-tag-driven schema generation.
For end-user installation and usage examples, see README.md and the guides in docs/.
Commands
Run commands from this directory.
task help # Show available task targets
task test # Run all tests with the race detector
task lint # Run golangci-lint and go mod tidy checks
task bench # Run benchmarks
task verify # Run deps, fmt, vet, lint, test, and govulncheck
Agent Operating Rules
- Read the relevant source, tests, and SPECS before changing behavior.
- Think through the user-visible contract and failure cases before coding.
- Apply KISS, DRY, and YAGNI; prefer the standard library and existing project utilities.
- Reproduce bugs through the affected public entry point before changing internals.
- Implement the smallest vertical slice that proves the requested behavior end to end.
- Keep edits surgical and preserve unrelated worktree changes.
- Verify against the task's actual acceptance criteria, not a generic command checklist.
- Load only the context needed for the current decision and follow references on demand.
- Resolve conflicting evidence in favor of runtime behavior, tests, and the owning SPEC; surface unresolved contract conflicts.
- Write behavior tests that explain the regression and fail loudly on unsupported states.
- Do not add policy-only gate scripts that merely restate documentation.
- Do not add spec-mirror tests when stronger public-behavior coverage already proves the contract.
Agent Workflow
Before designing or modifying behavior, identify and read the relevant document
in SPECS/. Confirm the current implementation and tests against that contract,
then make the smallest coherent change. If the code and SPEC disagree, resolve
which reflects the intended public behavior before editing either one.
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 · 180 lines · 2,208 tokens per session scan A 6eb9dc556723
jsonschema CLAUDE.md is an instructions file published in the GitHub repository kaptinlin/jsonschema (226 stars, last pushed 8d ago), licensed MIT. It adds 2,208 tokens to every session, about $0.0110 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 2026-08-30.
Other instructions, from other repositories
instructor CLAUDE.md
Claude Code instructions for 567-labs/instructor, covering claude.md, instructor development guide, commands, installation & setup and code style guidelines.
formkit CLAUDE.md
Instructions for formkit/formkit, covering formkit, architecture, packages, core (framework-agnostic) and vue integration.
mobx-formkit AGENTS.md
Instructions for foxhound87/mobx-formkit, covering guide for ai agents, 1. project identity, 2. philosophy, 3. architecture and supporting classes.
klaudiush CLAUDE.md
Claude Code instructions for smykla-skalski/klaudiush, covering claude.md, project overview, commands, completion (shell completion scripts) and doctor (diagnose setup and configuration).
klaudiush copilot-instructions.md
Copilot instructions for smykla-skalski/klaudiush, covering github copilot instructions, project overview, commands, build and development build (no signoff validation).
AgentRelay CLAUDE.md
Instructions for mnemox-ai/AgentRelay, covering agentrelay — claude code 指令, 專案概述, 架構（分層）, 開發規範 and tos 紅線.