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/prefect12/codex-token-meter/agents-mdgit clone --depth 1 https://github.com/prefect12/codex-token-meterWhat 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 | $0.01357 | $0.01357 |
| Opus 5 | $0.00678 | $0.00678 |
| Sonnet 5 | $0.00271 | $0.00271 |
| Haiku 4.5 | $0.00136 | $0.00136 |
Grade A, and why
codex-token-meter 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 — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Token Meter Development Guide
Project Shape
- This is a native macOS menu bar app built directly with
swiftc. - The app entrypoint lives in
Sources/CodexTokenMeter/main.swift; supporting code is split into focused files under the same directory. - The companion Task Bar app lives in
Sources/CodexPetBar/main.swiftand is part of the same product family, not a separate design surface. build.shcompiles every Swift file underSources/CodexTokenMeter, so future file splits do not need build-script changes.- Prefer small, behavior-preserving changes unless you are explicitly doing a planned refactor.
- Read
docs/ARCHITECTURE.mdbefore changing parser, quota, cost, or UI behavior.
Build And Verification
- Compile check:
./build.sh - CLI parser check:
"./build/AI Token Meter.app/Contents/MacOS/CodexTokenMeter" --print --window=week --quota=all - Live quota check:
"./build/AI Token Meter.app/Contents/MacOS/CodexTokenMeter" --print-live - Service-status check:
"./build/AI Token Meter.app/Contents/MacOS/CodexTokenMeter" --print-service-status - Dashboard render check for UI changes:
"./build/AI Token Meter.app/Contents/MacOS/CodexTokenMeter" --render-dashboard=/tmp/ai-token-meter-dashboard.png
--print-live, --print-profile, --print-service-status, and dashboard rendering can depend on Codex login state or network availability. Do not treat their unavailability as a compile regression unless the failure is caused by local code.
Git Workflow
- Prefer merging changes through pull requests. Do not merge directly into
mainunless the user explicitly asks for it. - All source, behavior, or UI changes must use this sequence: a focused branch → PR → merged
main→ fetchorigin/main→ build and install from that exact merged revision. Do not install an unmerged working-tree build as the delivered app. - Before calling a change complete, verify the PR merge, confirm local
HEADequalsorigin/main, then run the relevant checks and validate the installed/Applications/AI Token Meter.appsurface.
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 · 76 lines · 1,357 tokens per session scan A d77e8555cbe1
codex-token-meter AGENTS.md is an instructions file published in the GitHub repository prefect12/codex-token-meter (5 stars, last pushed 2d ago), licensed MIT. It adds 1,357 tokens to every session, about $0.0068 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
codex-switch AGENTS.md
AGENTS.md instructions for ChArLiEdance/codex-switch, covering repository agents and platform isolation.
codex-provider-sync AGENTS.md
Instructions for Dailin521/codex-provider-sync, covering ai / agent operator guide, vnext architecture baseline, goal, choose the interface and safe operating flow.
ai-code-interface.el AGENTS.md
Instructions for tninja/ai-code-interface.el, covering agents.md, common development commands, byte compilation, running tests and run a specific test file.
ai-code-interface.el copilot-instructions.md
Instructions for tninja/ai-code-interface.el, covering github copilot instructions for ai-code-interface.el, project overview, language and technology, code style and conventions and file naming.
droidproxy AGENTS.md
Instructions for nikships/droidproxy, covering agents.md, build & run, cli-proxy-api are stopped before the new app starts, debug build (no .app bundle, no relaunch) and build the .app bundle first, then open it.
sandvault AGENTS.md
Instructions for webcoyote/sandvault, covering agents.md and testing.