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/syhily/kobato/agents-mdgit clone --depth 1 https://github.com/syhily/kobatoWrote 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/syhily/kobato/agents-md)<a href="https://agentmods.dev/instructions/syhily/kobato/agents-md"><img src="https://agentmods.dev/badge/instructions/syhily/kobato/agents-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 | $0.07213 | $0.07213 |
| Opus 5 | $0.03606 | $0.03606 |
| Sonnet 5 | $0.01443 | $0.01443 |
| Haiku 4.5 | $0.00721 | $0.00721 |
Grade A, and why
kobato AGENTS.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 — 397 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Project conventions
Repository conventions for AI agents and contributors.
Quick orientation
- React Router 8 Framework Mode with SSR (
appDirectory: 'src'), React 19 TSX/TS only, SQLite (node:sqlite) + a DuckDB analytics sidecar. - Path alias
@/*→./src/*. - Five layers under
src/:routes/(orchestration),server/(SSR),client/(browser),ui/(components),shared/(isomorphic).
Config file
Infrastructure configuration lives in kobato.config.json — auto-created with defaults when missing. src/server/infra/config.ts's CONFIG_TABLE maps nested config paths to Zod schemas; the validated serverConfig object exported from that module is what consumers read. Env var names are derived by convention (path.join('__') → storage__database).
- Location order:
--config <path>→ SEA<execDir>/kobato.config.json→./kobato.config.json→~/.config/kobato.config.json. First existing wins. - Precedence: schema defaults < config file < env vars. Env values differing from the file are written back.
VITEST=truewithout--config→ env-only, zero filesystem access.- Tooling that spawns the binary MUST pass
--config <tempDir>/…to avoid persisting throwaway config next to the binary. - Adding a config value: add a
CONFIG_TABLErow → updatekobato.config.example.json→ cover it intests/unit/server/infra/config.test.ts.
Subdirectory conventions
Claude loads these additively as it moves through the codebase:
| File | Scope |
|---|---|
src/routes/AGENTS.md |
Route modules, loaders, actions, React Router conventions |
src/server/AGENTS.md |
Server layers (infra, domains, http, render), API procedures, auth |
src/client/AGENTS.md |
Browser hooks, oRPC client, React.lazy patterns |
src/ui/AGENTS.md |
Pure-props components, shadcn, PT renderer, component architecture |
src/styles/AGENTS.md |
Tailwind tokens, design-system CSS, @theme conventions |
src/shared/AGENTS.md |
Isomorphic modules, Zod contracts, DTOs, PT schema |
tests/AGENTS.md |
Test utilities, naming conventions, coverage rules |
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 · 397 lines · 7,213 tokens per session scan A 674fe61670c5
kobato AGENTS.md is an instructions file published in the GitHub repository syhily/kobato (115 stars, last pushed 3d ago), licensed MIT. It adds 7,213 tokens to every session, about $0.0361 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
repo-remover CLAUDE.md
Claude Code instructions for moollaza/repo-remover, covering repo remover, stack, constraints, commands and commits.
tanstack-hono AGENTS.md
Instructions for bskimball/tanstack-hono, covering ai agent guidelines, essentials, guides, using vite+, the unified toolchain for the web and vite+ workflow.
tanstack-hono CLAUDE.md
Instructions for bskimball/tanstack-hono, covering claude guidelines, quick reference, using vite+, the unified toolchain for the web, vite+ workflow and start.
rockpack CLAUDE.md
Claude Code instructions for AlexSergey/rockpack, covering project guidelines, important, typescript rules, testing rules and test structure.
repo-remover AGENTS.md
AGENTS.md instructions for moollaza/repo-remover, a project described as: A web app that helps you archive and delete old/unused repos, quickly and easily.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).