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/svd-ai-lab/sim-cli/claude-mdgit clone --depth 1 https://github.com/svd-ai-lab/sim-cliWhat 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 | $0.04423 | $0.04423 |
| Opus 5 | $0.02211 | $0.02211 |
| Sonnet 5 | $0.00885 | $0.00885 |
| Haiku 4.5 | $0.00442 | $0.00442 |
Grade A, and why
sim-cli CLAUDE.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 — 239 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Project Overview
sim reads existing CAE files and turns them into structured text an agent can use, detects local solver installs, validates scripts, and can hold a live solver session open across steps. The pyproject.toml description — "Agent-native simulation asset parsing and CAD/CAE runtime" — states the priority correctly: parsing first, runtime second.
Know which tier you are working in, because their dependencies differ:
- Asset parsing (
sim scan) — pure core, no plugin, no solver, no license. Backed by thesimparsenative extension, a hard dependency ofsim-cli-core. This is the most-used capability in practice: agents can already drive solver GUIs directly, so the highest-value thingsimdoes today is recover engineering context from historical case files (a preprocessing step, closer to a document-to-structured-text converter than to a solver driver). - Solver discovery and script validation (
sim check,sim lint) — require the solver's plugin. With no plugins installed,check --allreports "scanned 0 drivers" andlinterrors withNo registered driver claims .... - Execution — requires a plugin, and comes in two modes:
- One-shot (
sim run script --solver=X): subprocess execution, result stored as a numbered run,sim logsto browse. - Persistent session (
sim connect/exec/inspect/disconnect): a long-lived HTTP server holds a live solver session; agents send code snippets and inspect state without restarting the solver.sim connectauto-starts a local server when none is reachable (session.py:_auto_start_server), sosim serveis only needed explicitly for remote hosts — andsim stopis the counterpart that reaps that background process.
- One-shot (
This tiering is load-bearing for docs and error messages: never imply sim scan needs a plugin, and never imply sim check works without one.
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 · 239 lines · 4,423 tokens per session scan A f265452b8fbb
sim-cli CLAUDE.md is an instructions file published in the GitHub repository svd-ai-lab/sim-cli (213 stars, last pushed 5d ago), licensed Apache-2.0. It adds 4,423 tokens to every session, about $0.0221 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.
Other instructions, from other repositories
awesome-ai-cae CLAUDE.md
Instructions for kimimgo/awesome-ai-cae, covering claude.md, project info, ci — lint & link check, awesome-lint: readme.md 포맷 검증 (push/pr 시 자동 실행) and lychee: 링크 깨짐 검사 (push/pr + 매주 월요일 06:00 utc).
chrono AGENTS.md
AGENTS.md instructions for projectchrono/chrono, covering repository guidelines, developers vs. users, project structure & module organization, build, test, and development commands and physics manuals & demo-first workflow.
Foam-Agent AGENTS.md
Instructions for csml-rpi/Foam-Agent, covering agents.md, what is foam-agent?, build and run, environment setup and run a simulation.
cad-cae-copilot AGENTS.md
Instructions for armpro24-blip/cad-cae-copilot, covering aieng workbench — agent guide, stop — read this first, first three calls every session, workflow priority matrix and workspace layout.
omnisim AGENTS.md
Instructions for omnilink-tech/omnisim, covering agents.md — running omnisim as an ai coding agent, 0. for agents: read this first (60 seconds), run this first turn, first moves by task type and hard-won rules (don't relearn these).
viznoir CLAUDE.md
Instructions for kimimgo/viznoir, covering claude.md, viznoir, project info, development commands and install (editable with dev deps).