agentmods.dev

ai-usage-widget AGENTS.md

Project instructions for building a desktop widget that shows local usage of AI coding tools. Tauri is a framework for making desktop apps with a web-based interface and a small native shell.

In plain words
What is it for?
Use them to build provider detection, safe command execution, data parsers and tests, a shared usage model, refresh scheduling, and the compact widget interface.
Why use it?
They give the agent the project goals and constraints, including local-only data collection, support for several operating systems, and keeping the interface responsive.

Instructions file for CodexOpenCode

7 7d ago A scan 1,575 tokens original MIT
View the source file odrasile/ai-usage-widget
Install

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.

agentmods
npx agentmods add instructions/odrasile/ai-usage-widget/agents-md
Clone the repo
git clone --depth 1 https://github.com/odrasile/ai-usage-widget

Made for: Codex, OpenCode.

Per session 1,575 This file is loaded in full into every session.
When invoked 1,575 The same file — it is already loaded in full.
Security scan A 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.01575 $0.01575
Opus 5 $0.00788 $0.00788
Sonnet 5 $0.00315 $0.00315
Haiku 4.5 $0.00158 $0.00158

Measured 2d ago against content hash 2bd3f79e2039, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ai-usage-widget AGENTS.md scanned grade A with 1 finding 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.

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

- CLI execution: `child_process` and, when a real TTY is required, a pseudo-terminal.
AGENTS.md · 251 lines

How it starts

The opening of the file, as written. The whole thing — 251 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

Goal

Implement and maintain a desktop widget for Windows, Ubuntu Desktop, and macOS that monitors AI coding tool usage through local CLIs.

The expected result is:

  • A functional Tauri desktop application.
  • A TypeScript frontend.
  • A compact, floating, always-on-top, borderless widget.
  • Local integration with Codex, Claude Code, and Gemini.
  • Working builds for Windows, Ubuntu Desktop, and macOS.
  • A simple, modular base that is ready to distribute.

Working Principles

  1. KISS: keep the code simple and direct.
  2. Avoid overengineering.
  3. Separate responsibilities by module.
  4. Do not add dependencies unless they solve a real limitation.
  5. Do not use the network or external APIs to obtain usage data.
  6. Prioritize fault tolerance: if one provider fails, the widget must keep working.
  7. Do not block the UI while querying CLIs.

Workflow

  1. Initialize or review the Tauri + frontend project.
  2. Implement CLI detection.
  3. Implement safe command execution.
  4. Implement provider adapters.
  5. Implement parsers with tests.
  6. Implement the unified data model.
  7. Implement the minimal UI.
  8. Integrate the refresh scheduler.
  9. Validate detection, parsing, and UI behavior.
  10. Prepare builds and installers for Windows, Ubuntu Desktop, and macOS.
  11. Document the result.

Do not skip phases unless the repository already has that phase completed.

Technical Constraints

  • Frontend: TypeScript.
  • Desktop framework: Tauri.
  • Local backend: Node.
  • CLI execution: child_process and, when a real TTY is required, a pseudo-terminal.
  • Do not use Electron.
  • Do not use heavy UI libraries.
  • Do not use external APIs to query usage.
  • Do not implement history, multi-user support, or browser extensions.

Codex Integration

Codex must not be queried with codex status, because current versions do not expose status as a subcommand.

Correct flow:

  1. Detect installation with where.exe codex on Windows or which codex on Unix/macOS.
  2. Open codex --no-alt-screen inside a pseudo-terminal.
  3. Send /status.
  4. Capture output.
  5. Exit with /quit.

Read the full file on GitHub · 251 lines

Changes

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.

  1. 2d ago First seen · 251 lines · 1,575 tokens per session scan A 2bd3f79e2039

Subscribe to this mod's changes

ai-usage-widget AGENTS.md is an instructions file published in the GitHub repository odrasile/ai-usage-widget (7 stars, last pushed 7d ago), licensed MIT. It adds 1,575 tokens to every session, about $0.0079 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Analytics and metricsai-usageclaude-codeclaude-usagecodexcodex-clicross-platform
Related

Other instructions, from other repositories

whipnext AGENTS.md

Click a pack character overlay; it plays hit media and types a phrase (default next) into the focused coding-agent session.

letya999/whipnext · 1,165 tokens

codex-discord-presence AGENTS.md

AGENTS.md instructions for dayfinggg/codex-discord-presence, covering codex discord presence, code map and commands.

dayfinggg/codex-discord-presence · 562 tokens

Codex_Relay AGENTS.md

AGENTS.md instructions for Red-noblue/Codex_Relay, a project described as: Cross-device export/import + local vault for Codex CLI sessions (.jsonl). Resume the same session on macOS/Windows/Manage, backup, and transfer Codex CLI sessions (.jsonl) across devices with preview + conflict-safe import.Linux.

Red-noblue/Codex_Relay · not measured

apm python.instructions.md

Python development guidelines.

microsoft/apm · 78 tokens

moonproxy-desktop AGENTS.md

Instructions for MoonProxyHQ/moonproxy-desktop, covering moonproxy, 功能, 技术栈, 目录结构 and 配置与数据存储.

MoonProxyHQ/moonproxy-desktop · 3,883 tokens

claude-code-profiles CLAUDE.md

Instructions for quinnjr/claude-code-profiles, covering claude.md, project, architecture, command interface and directory-local profiles.

quinnjr/claude-code-profiles · 1,863 tokens