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/vincentkoc/tokenjuice/agents-mdgit clone --depth 1 https://github.com/vincentkoc/tokenjuiceWrote 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/vincentkoc/tokenjuice/agents-md)<a href="https://agentmods.dev/instructions/vincentkoc/tokenjuice/agents-md"><img src="https://agentmods.dev/badge/instructions/vincentkoc/tokenjuice/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 | $0.01397 | $0.01397 |
| Opus 5 | $0.00698 | $0.00698 |
| Sonnet 5 | $0.00279 | $0.00279 |
| Haiku 4.5 | $0.00140 | $0.00140 |
Grade A, and why
tokenjuice 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 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.
How it starts
The opening of the file, as written. The whole thing — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository Guidelines
Project Structure & Module Organization
tokenjuice is a TypeScript CLI/library. Main source lives in src/:
src/cli/contains the CLI entrypoint.src/core/contains shared reducers, rule loading, and other host-agnostic logic.src/hosts/contains host-specific integrations (codex/,claude-code/,cursor/,pi/,opencode/,openclaw/) plus shared hook helpers undershared/.src/rules/contains built-in JSON rules and fixtures.
Tests live in test/ and mirror the source layout (test/core/, test/hosts/, test/hosts/shared/). Supporting docs live in docs/, packaging files in packaging/, and utility scripts in scripts/. dist/, release/, and src/core/builtin-rules.generated.ts are generated outputs; do not hand-edit them.
Build, Test, and Development Commands
pnpm installinstalls dependencies.pnpm testruns the full Vitest suite and regenerates built-in rules first.pnpm typecheckruns TypeScript without emitting files.pnpm buildrebuildsdist/, Pi runtime output, and packaged rules.pnpm verifyis the pre-handoff gate: lint, circular-dep check (madge), typecheck, and the full Vitest suite.pnpm release:localruns the local release pipeline: tests, build, tarball, checksums, and Homebrew formula.pnpm exec vitest run test/hosts/codex.test.tsruns a focused test file during iteration.
For a quick smoke check after building, run node dist/cli/main.js --version.
Coding Style & Naming Conventions
Use TypeScript with ESM imports and explicit .js import suffixes in source files. Match the existing style:
- 2-space indentation
- named exports over default exports for library modules
- small, focused helpers in
src/core/ - descriptive test names using
describe/it
Keep generated files out of manual edits. If you change built-in rule data, rely on the existing scripts and test/build flows to regenerate derived output.
Host integrations
Every post-tool host adapter (codex, pi, opencode, openclaw) routes through the shared primitive compactBashResult in src/core/integrations/compact-bash-result.ts. New host behavior belongs in src/core/, not in an adapter. Pre-tool rewriters (claude-code, codebuddy, cursor, vscode-copilot) funnel commands through tokenjuice wrap and shared helpers in src/hosts/shared/pre-tool-wrap.ts. When adding or changing a host, see docs/integration-playbook.md for the install/doctor/uninstall checklist and update the hardcoded host list in src/cli/main.ts accordingly.
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.
- 4d ago First seen · 83 lines · 1,397 tokens per session scan A 357f4884a92f
tokenjuice AGENTS.md is an instructions file published in the GitHub repository vincentkoc/tokenjuice (514 stars, last pushed 2mo ago), licensed MIT. It adds 1,397 tokens to every session, about $0.0070 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
ClawTeam-OpenClaw CLAUDE.md
Instructions for win4r/ClawTeam-OpenClaw, covering claude.md, what this project is, common commands, dev install (from repo root) and cli entry points.
hex1b AGENTS.md
Instructions for mitchdenny/hex1b, covering ai coding agent guidelines for hex1b, 🛠️ available skills, 📋 project overview, key technologies and repository layout.
toofan AGENTS.md
Instructions for vyrx-dev/toofan, covering agents.md, build & run, project structure, architecture rules and adding content.
ClimaAtmos.jl AGENTS.md
Instructions for CliMA/ClimaAtmos.jl, covering climaatmos agent guide, ecosystem guidelines, repo-specific guidelines, local norms and self-correction.
Claude-Code-Promts-Skills CLAUDE.md
Instructions for Rtur2003/Claude-Code-Promts-Skills, covering claude.md, project overview, repository structure, file naming conventions and prompt structure standards.
vibesolve AGENTS.md
Instructions for vibesolve/vibesolve, covering agents.md, what this project is, setup (one-time), cli and pipeline architecture.