What is cosim-gpu-test?

A testing procedure for running GPU programs in cosim, a GPU simulation environment, and classifying their results. It creates test evidence, checks that the requested program is the right one, and cleans up normal test sessions.

Use it to run named GPU tests or a full test set, classify outcomes, preserve artifacts, and inspect a manually launched cosim session when needed.

It makes GPU test results reproducible and traceable instead of relying on an interactive run with unclear inputs or missing evidence. It also distinguishes ordinary failures from hangs and environment problems.

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-test/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-test/github.svg)](https://agentmods.dev/skills/gevico/cosim-gpu/cosim-gpu-test)

Your own site

<a href="https://agentmods.dev/skills/gevico/cosim-gpu/cosim-gpu-test"><img src="https://agentmods.dev/badge/skills/gevico/cosim-gpu/cosim-gpu-test/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-test"><img src="https://agentmods.dev/badge/skills/gevico/cosim-gpu/cosim-gpu-test.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 841 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	$0.00841
Opus 5.5	$0.00018	$0.00336
Sonnet 5.5	$0.00009	$0.00168
Haiku 4.5	$0.00005	$0.00084

Measured 11d ago against content hash 8f3e434f1e51, 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-test 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 11d 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-test/SKILL.md · 92 lines

How it starts

The opening of the file, as written. The whole thing — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Cosim Test

Execute GPU programs against cosim in fresh sessions. The test runner owns launching, vHost execution, exit status, failure categories, diagnostic logs, and session cleanup. Record build provenance through cosim-gpu-build.

Use cosim-gpu-launch only for manual boot or interactive inspection. Use this skill whenever a program result, matrix row, verification result, or PR-grade test artifact is required.

Quickstart

just test-gpu vector_add
just test-gpu --all
just test-gpu --repeat 3 vector_add
just test-gpu --test-timeout 120 vector_add
just test-gpu --keep-alive vector_add

just test-gpu runs one program in a fresh cosim session. --all and --repeat N start a fresh session for every program or repetition. Use --keep-alive when a successful run must leave the vHost available for manual inspection; use just run-launch for a manual vHost session.

Set COSIM_RUN_ID to make a run addressable. On failure, the runner records the run id, category, program and exit code with the captured vHost and simulator logs.

Required Gates

Before launching a named program, prove exact program identity. Read references/program-identity-and-vhost-env.md for local kernels, ROCm examples, variants, timeout estimation, device-side printf, or GUEST_TEST_PREFIX.

Before automated rows, repeats, environment matrices, non-default binaries, or evidence intended to prove uncommitted source, read references/run-manifest-and-provenance.md.

For long diagnostic rows, performance rows, matrix runs, or interrupted test automation, read references/diagnostic-and-matrix-runs.md.

Use cosim-gpu-build for build readiness and binary provenance decisions. Do not duplicate build metadata rules here.

Automation Rule

When program identity, manifest rows, and required provenance checks are complete, run the just test-gpu recipe. Do not ask before standard test execution, repeats, interrupt-mode matrix rows, timeout probes, post-fix reruns inside the current scope, standard incremental build, or standard preflight cleanup.

Read the full file on GitHub · 92 lines

Files

What ships with it

4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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.

11d ago Changed 8f3e434f1e51
12d ago Changed · -9 lines · -3 tokens per session c69bc6e480c2
19d ago First seen · 101 lines · 48 tokens per session scan A 38460488f049

Subscribe to this mod's changes

cosim-gpu-test 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 841 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.

Testing

research-engineer

An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.

davila7/claude-code-templates · 43 tokens

tika-eval-compare

Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".

apache/tika · 50 tokens

cua-driver

Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.

davidondrej/skills · 68 tokens

cosim-gpu: Skill for Claude Code