model-usage AGENTS.md

Project instructions for a command-line tool that reads local usage data from AI coding assistants and summarizes model use and costs. It covers the tool's structure, data sources, filters, sorting, pricing, and project detection.

In plain words
What is it for?
Use it when working on features such as source selection, project or date filtering, cost and token summaries, model pricing, or terminal display.
Why use it?
When several coding assistants store usage data in different places, it is difficult to compare activity and spending in one view. These instructions explain how the project finds and organizes that data.

Instructions file for CodexOpenCode

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/evanlong-me/model-usage/agents-md
Clone the repo
git clone --depth 1 https://github.com/evanlong-me/model-usage

Made for: Codex, OpenCode.

Per session 1,868 This file is loaded in full into every session.
When invoked 1,868 The same file — it is already loaded in full.
Security scan A 0 findings. 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 $0.01868 $0.01868
Opus 5 $0.00934 $0.00934
Sonnet 5 $0.00374 $0.00374
Haiku 4.5 $0.00187 $0.00187

Measured 2d ago against content hash a0e2ab654450, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

model-usage 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.

AGENTS.md · 180 lines

How it starts

The opening of the file, as written. The whole thing — 180 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Model Usage

Zero-config CLI for analyzing AI model usage & costs across all local TUI tools — pi, Claude Code, Codex, Gemini CLI, OpenCode, and more.

Project Structure

├── src/
│   ├── cli.ts               # CLI entry point (Commander setup)
│   ├── display.ts            # Output formatting (tables, models, projects)
│   ├── usage.ts              # Read usage data (auto-detect all sources)
│   ├── aggregator.ts         # Aggregate messages by project/date
│   ├── filters.ts            # Time and project filtering
│   ├── sorter.ts             # Sorting by cost/time/tokens/project
│   ├── pricing.ts            # Fetch model pricing from LiteLLM
│   ├── project-detector.ts   # Auto-detect current project
│   ├── update-checker.ts     # Check for npm updates
│   ├── github-prompt.ts      # GitHub star prompt
│   ├── source-selector.ts    # Interactive TUI source picker
│   ├── util.ts               # Shared utilities (path, fetch)
│   ├── types.ts              # TypeScript type definitions
│   ├── cli-table3.d.ts       # Type declarations for cli-table3
│   └── sources/              # Auto-discovered source adapters
│       ├── index.ts          # Auto-discovery (no manual imports needed)
│       ├── common.ts         # Shared utilities for all sources
│       ├── pi.ts             # pi sessions
│       ├── claude.ts         # Claude Code sessions
│       ├── codex.ts          # Codex CLI sessions
│       ├── gemini.ts         # Gemini CLI sessions
│       ├── opencode.ts       # OpenCode sessions (SQLite)
│       └── grok.ts           # Grok CLI sessions (SQLite)
├── dist/                     # Compiled output (gitignored)
├── tsconfig.json
└── package.json

Key Commands

# Default (interactive source selection if TTY, else all sources)
mu                     # Show all usage for auto-detected project
mu -a                  # Show all projects (skip auto-detection)

# Source selection
mu -s pi,opencode      # Only query specific TUI tools (long form: --sources)
mu --sources all          # Query all sources (skip selection)
mu                        # Interactive checkbox UI to pick sources

# Filtering
mu -t 7d               # Last 7 days (rolling, relative to "now"); also 30min, 2h, 1m, 1y
mu -t today            # Calendar keywords: today, yesterday, thisweek, lastweek,
                       # thismonth, lastmonth, thisyear, lastyear (week starts Monday;
                       # hyphenated aliases like this-week work too)
mu -t 7-8              # Month range this year; july-august or 2025-7-2026-8 also work
mu -t 2026-07-25       # A single date — that whole calendar day
mu -p <project>        # Filter by project (partial match: "model", "obix")
mu -m <model>          # Filter by model (partial match: "sonnet", "gpt-5")

# Sorting
mu -k cost -o desc     # Sort by cost descending
mu -k time -o asc      # Sort by time ascending (default)

# View modes
mu -d                  # Detailed view (individual messages)
mu -b                  # Aggregate by date only (combines all projects)

# Lists
mu -P                  # List projects
mu -M                  # List models with pricing

# Combined
mu -m sonnet -t 7d     # Last 7 days, sonnet models only
mu -m haiku -p myproject --by-date  # Per-day, haiku only, specific project

Read the full file on GitHub · 180 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 · 180 lines · 1,868 tokens per session scan A a0e2ab654450

Subscribe to this mod's changes

model-usage AGENTS.md is an instructions file published in the GitHub repository evanlong-me/model-usage (21 stars, last pushed 27d ago), licensed MIT. It adds 1,868 tokens to every session, about $0.0093 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.