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/choi138/toki/agents-mdgit clone --depth 1 https://github.com/choi138/tokiWhat 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 | $0.01581 | $0.01581 |
| Opus 5 | $0.00790 | $0.00790 |
| Sonnet 5 | $0.00316 | $0.00316 |
| Haiku 4.5 | $0.00158 | $0.00158 |
Grade A, and why
toki 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 2d 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 — 141 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Workspace Rules
This file is the high-level policy entry point for AI tooling in this repository.
Keep detailed source rules out of AGENTS.md; store them in the Codex conventions
files referenced below.
Source Of Truth
| Topic | Source | When |
|---|---|---|
| Toki Source Conventions | /project-conventions skill or .agents/skills/project-conventions/conventions.md plus task-specific files under .agents/skills/project-conventions/references/ |
Before editing menubar/Toki/**/*.swift, menubar/TokiTests/**/*.swift, menubar/project.yml, .swiftformat, .swiftlint.yml, or app resources |
| Local Codex Review Loop | /codex-review-loop skill or .agents/skills/codex-review-loop/SKILL.md |
Local diff review, approved finding fixes, and bounded re-review |
| Git Workflow | .agents/conventions/git-workflow.md |
Branch / commit / PR work |
The shared .agents hooks provide soft enforcement for /project-conventions
at session start, on matching prompts, and on Bash commands that appear to
modify Toki source, tests, resources, or XcodeGen configuration.
This is a Codex-native reminder layer, not a hard block for non-shell edits.
Toki Workflow
Toki is a macOS menu bar app built with Swift, SwiftUI, XcodeGen, SwiftFormat, and SwiftLint.
- Inspect the existing local pattern before changing code.
- Keep app entry code in
menubar/Toki/App, domain models and builders inmenubar/Toki/Domain, platform and data readers inmenubar/Toki/Infrastructure, feature UI/view-model code inmenubar/Toki/Features, and tests inmenubar/TokiTests. - Prefer editing
menubar/project.ymland regenerating withxcodegen generatefrommenubar/for project configuration changes. Avoid hand-editingmenubar/Toki.xcodeprojunless the user explicitly asks for a project-file-only fix or regeneration is not possible. - Treat local agent logs, usage databases, security audit findings, and secret detection data as sensitive. Do not add telemetry, network transmission, or unmasked logging for those values unless explicitly requested.
- Keep SwiftUI views focused on presentation. Move parsing, aggregation, pricing, security scanning, persistence, and time calculations into domain, infrastructure, or view-model code.
- Add or update focused tests for reader behavior, usage aggregation, formatting, security audit behavior, settings, and view-model logic when those surfaces change.
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.
- 2d ago First seen · 141 lines · 1,581 tokens per session scan A d63cee85a600
toki AGENTS.md is an instructions file published in the GitHub repository choi138/toki (9 stars, last pushed 2d ago), licensed MIT. It adds 1,581 tokens to every session, about $0.0079 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.
Other instructions, from other repositories
ai AGENTS.md
Instructions for team-telnyx/ai, covering agents.md, what this repo is, working on this repo, setup and root tooling (only for npm run test:guides).
tokentelemetry AGENTS.md
AGENTS.md instructions for VasiHemanth/tokentelemetry, covering agents.md, project, common commands and project layout.
Ocul-PM CLAUDE.md
Instructions for bunhine0452/Ocul-PM, covering claude.md, what this is, commands, single frontend test and rust (run from src-tauri/).
Ocul-PM AGENTS.md
Instructions for bunhine0452/Ocul-PM, covering ocul-pm 작업 기록 규칙, 0. 시작하기 전 — 과거를 먼저 찾는다, 1. 언제 기록하는가 (5 trigger), 2. 일지 파일 규격 (도구가 없을 때만) and 3. 금지.
rrweb-stack AGENTS.md
Instructions for Cubenest/rrweb-stack, covering project context for ai coding assistants, what the repo is, build, test & dev commands, planning context — the context/ convention and maintainer-only setup (do this once per fresh checkout).
agenttop CLAUDE.md
Instructions for vicarious11/agenttop, covering agenttop — ai usage dashboard & optimizer, project overview, architecture, layers and key directories.