What is cosim-gpu-launch?

A command-line launcher for a QEMU and gem5 co-simulation, where QEMU runs a virtual machine and gem5 models hardware behavior. It starts the environment for manual booting, inspection, and standalone sessions.

Use it to start the GPU co-simulation environment, choose its communication backend, enable gem5 debugging, and inspect a running guest system.

It removes the need to assemble and start the simulator components by hand. It is not a test runner and does not create managed test evidence.

How gevico/cosim-gpu configures Claude Code and Codex

Reuse

Borrowing it

Nothing to install: this file belongs to gevico/cosim-gpu. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/gevico/cosim-gpu/main/.agents/skills/cosim-gpu-launch/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/gevico/cosim-gpu

Made for: Claude Code, Codex.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/gevico/cosim-gpu/cosim-gpu-launch/github.svg)](https://agentmods.dev/skills/gevico/cosim-gpu/cosim-gpu-launch)

Your own site

<a href="https://agentmods.dev/skills/gevico/cosim-gpu/cosim-gpu-launch"><img src="https://agentmods.dev/badge/skills/gevico/cosim-gpu/cosim-gpu-launch/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/gevico/cosim-gpu/cosim-gpu-launch"><img src="https://agentmods.dev/badge/skills/gevico/cosim-gpu/cosim-gpu-launch.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 45 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 804 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00045	$0.00804
Opus 5.5	$0.00018	$0.00322
Sonnet 5.5	$0.00009	$0.00161
Haiku 4.5	$0.00005	$0.00080

Measured 4d ago against content hash e9ee2ae86977, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

cosim-gpu-launch 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 4d 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/skills/cosim-gpu-launch/SKILL.md · 84 lines

How it starts

The opening of the file, as written. The whole thing — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Cosim Launch

Launch the QEMU+gem5 co-simulation environment.

This is a direct launcher, not the test harness. Use it for manual boot, interactive inspection, backend checks, and standalone QEMU+gem5 sessions. It does not accept --output-dir, does not create test matrices, does not write vHost program logs, and does not run binary-provenance hooks. Use cosim-gpu-test when a run must produce case-managed evidence.

Quickstart

just run-preflight
just run-launch
just run-launch --gem5-debug MI300XCosim
just run-launch --cosim-backend legacy

After the vHost boots, the driver loads automatically via cosim-gpu-initialize.service. If not, manually load with cosim-gpu-guest.

Autonomy

Run just run-preflight to inspect Host resources before launch. It does not change files or remove resources. Use just run-clean-orphans only when orphaned resources need removal.

Backends

Backend	Protocol	Default?	Notes
vfio-user	`libvfio-user` over Unix socket	Yes	Standard, no custom QEMU code
legacy	Custom cosim socket protocol	No	QEMU `mi300x-gem5` device

Architecture

flowchart LR
    subgraph H[Host OS]
        Q[Host-side QEMU]
        subgraph C[Docker container]
            D[gem5 MI300X Device Model]
        end
    end
    Q -->|KVM or TCG| V[vHost OS]
    Q <-->|vfio-user / libvfio-user| D

Shared memory uses run-id names: /dev/shm/cosim-guest-ram-<run-id> for vHost RAM and /dev/shm/mi300x-vram-<run-id> for VRAM. BAR layout: 0+1=VRAM, 2+3=Doorbell, 4=MSI-X, 5=MMIO.

Driver parameters

ip_block_mask=0x67  # enable common/GMC/IH/GFX/SDMA; disable PSP bit 3 and SMU bit 4
ppfeaturemask=0     # disable power play
dpm=0               # disable dynamic power management
audio=0             # no audio controller
ras_enable=0        # disable RAS
discovery=2         # firmware-based discovery

Troubleshooting

Symptom	Route
gem5 container exited during launch	Use `cosim-gpu-debug` to inspect active gem5 container logs
Driver failed while parsing AtomBIOS	Use `cosim-gpu-guest` or launch setup workflow to inspect ROM visibility before amdgpu loads
KIQ disable timeout (-110)	Treat as expected cosim noise unless followed by a functional failure
DRM client -13 / EPERM	Use vHost setup or disk-image workflow to inspect permissions and image state

Read the full file on GitHub · 84 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.

4d ago Changed e9ee2ae86977
11d ago Changed · +8 lines · -2 tokens per session 639d3357b660
12d ago Changed · -2 lines 625523ee1d9f
19d ago First seen · 78 lines · 47 tokens per session scan A 56af64bb125c

Subscribe to this mod's changes

cosim-gpu-launch is a skill published in the GitHub repository gevico/cosim-gpu (95 stars, last pushed 4d ago), licensed Apache-2.0. It adds 45 tokens to every session and 804 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-10-10.

Containers DevOps and cloud Hardware and IoT

holoscan-install-container

Install and verify the Holoscan SDK 5.x CUDA 13 development container after checking release availability and host compatibility.

NVIDIA/skills · 30 tokens

holoscan-install-source

Build and verify Holoscan SDK 5.x from an explicit source ref with the repository run wrapper; use for Early Access or SDK development.

NVIDIA/skills · 35 tokens

doca-urom-svc

Operate the DOCA UROM Service container on BlueField Arm for remote memory operations (puts, gets, atomics, collectives) enqueued by a paired host using doca-urom: pull the NGC image, choose the UCX component, size queues, configure Comch pairing, and align host and service versions. SECURITY: the service has no…

NVIDIA/skills · 153 tokens

doca-dts

Deploy and configure the NVIDIA DOCA Telemetry Service (DTS) as it is actually run: (A) BlueField (built-in static pod), (B) Host (docker) on x8664 / aarch64, or (C) DPF (DPUService / helm chart) — plus the shared configuration layer (dtsconfig.ini, providers, exporters, level labels) that all three share. The user's…

NVIDIA/skills · 240 tokens

hsb-setup

Clone the latest NVIDIA Holoscan Sensor Bridge repo, ask which supported devkit is being used, configure the host per platform, build the correct demo container, run it, and verify HSB connectivity by pinging 192.168.0.2. Use for Holoscan Sensor Bridge setup, build, container launch, and first-connectivity bring-up.

NVIDIA/skills · 78 tokens

routeros-container

RouterOS /container subsystem for running OCI containers on MikroTik devices. Use when: enabling containers on RouterOS, setting up VETH/bridge networking for containers, managing container lifecycle via CLI or REST API, building OCI images for RouterOS, configuring container environment variables, troubleshooting…

tikoci/routeros-skills · 86 tokens

cosim-gpu: Skill for Claude Code