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/metrillm/metrillm/agents-mdgit clone --depth 1 https://github.com/MetriLLM/metrillmWrote 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/metrillm/metrillm/agents-md)<a href="https://agentmods.dev/instructions/metrillm/metrillm/agents-md"><img src="https://agentmods.dev/badge/instructions/metrillm/metrillm/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 | $0.01187 | $0.01187 |
| Opus 5 | $0.00593 | $0.00593 |
| Sonnet 5 | $0.00237 | $0.00237 |
| Haiku 4.5 | $0.00119 | $0.00119 |
Grade A, and why
metrillm 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 3d 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 — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — MetriLLM CLI (MetriLLM/metrillm)
Guidelines for AI agents and contributors working on this repository.
Project Overview
MetriLLM is an open-source CLI tool that benchmarks local LLM models running on Ollama or LM Studio. It measures performance (tok/s, TTFT, memory, CPU load) and quality (reasoning, math, coding, instruction following, structured output, multilingual), then computes a hardware fitness verdict.
- License: Apache 2.0
- Language: TypeScript (ESM, Node 20+)
- Runtimes: Ollama (
src/core/ollama-client.ts), LM Studio (src/core/lm-studio-client.ts), abstracted viasrc/core/runtime.ts - Companion repo:
MetriLLM/metrillm-web(private) — leaderboard website atmetrillm.dev
Project Structure
src/
benchmarks/ # Benchmark runners (performance + 6 quality categories)
commands/ # CLI command handlers (bench, list)
core/ # Infrastructure (Ollama client, hardware detection, storage, upload, telemetry)
datasets/ # Ground truth JSON fixtures for quality evaluation
scoring/ # Score computation (performance, quality, fitness verdict)
ui/ # CLI output (tables, spinners, menus, verdict display)
index.ts # Entry point — Commander.js CLI with interactive menu fallback
types.ts # Shared types (also used by companion website)
tests/ # Vitest unit/regression tests (mirrors src/ structure)
mcp/ # MCP server for IDE integration (Claude Code, Cursor, etc.)
plugins/ # IDE plugins (Claude Code, Cursor)
scripts/ # Smoke/e2e helpers
docs/ # Technical documentation
Commands
npm run dev # Run CLI from source (tsx)
npm run build # Build distributable CLI (tsup → dist/index.mjs)
npm run typecheck # tsc --noEmit
npm test # Vitest run (all tests)
npm run test:watch # Vitest watch mode
npm run test:coverage # Vitest with V8 coverage (thresholds enforced)
npm run ci:verify # Full CI gate: typecheck + coverage + build
npm run security:audit # Dependency vulnerability audit
npm run test:e2e:smoke # Real Ollama integration test (requires running Ollama)
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.
- 3d ago First seen · 100 lines · 1,187 tokens per session scan A 5d22df492f77
metrillm AGENTS.md is an instructions file published in the GitHub repository MetriLLM/metrillm (5 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 1,187 tokens to every session, about $0.0059 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
dotty-stackchan CLAUDE.md
Instructions for BrettKinny/dotty-stackchan, covering dotty, ai transparency (binding on agents), what this is, architecture and network.
aws-bench AGENTS.md
Instructions for aws-bench/aws-bench, covering agents.md, setup, commands, project structure and code style.
brilliant_sdk AGENTS.md
Instructions for brilliantlabsAR/brilliant_sdk, covering brilliant sdk — agent guide, how an app works (the pattern behind everything), minimal reading paths, verify without hardware and testing.
NeoMind CLAUDE.md
Instructions for camthink-ai/NeoMind, covering neomind — edge ai platform for iot, development commands, ecosystem repositories, extension package contract (.nep) and device type template contract (json).
AMD-BC-250-Windows-Driver AGENTS.md
Instructions for Keshas-dev/AMD-BC-250-Windows-Driver, covering amd bc-250 windows driver — agent notes, host machine, ⚠️ historical-data caution (2026-08-21, strengthened), fresh re-verifications (current build, 2026-08-21) and workspace boundary.
Nothan-OS CLAUDE.md
Instructions for Vinalinux-Org/Nothan-OS, covering claude.md — nothanos, hard rules, toolchain, driver and design philosophy.