A set of instructions for choosing and running focused tests for OpenClaw Machines changes. It covers backend, frontend, workers, host enrollment, Firecracker, root filesystems, and self-hosted deployments.

Use it before commits, pushes, or pull requests to select tests for the affected code and rerun the smallest useful verification.

It helps verify the part of the system that changed without automatically running every expensive test. It also guides debugging and deciding when broader checks are needed.

How mathaix/OpenClawMachines configures Claude Code and Codex

Reuse

Borrowing it

Nothing to install: this file belongs to mathaix/OpenClawMachines. 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/mathaix/OpenClawMachines/main/.agents/skills/ocm-testing/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/mathaix/OpenClawMachines

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/mathaix/openclawmachines/ocm-testing/github.svg)](https://agentmods.dev/skills/mathaix/openclawmachines/ocm-testing)

Your own site

<a href="https://agentmods.dev/skills/mathaix/openclawmachines/ocm-testing"><img src="https://agentmods.dev/badge/skills/mathaix/openclawmachines/ocm-testing/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/mathaix/openclawmachines/ocm-testing"><img src="https://agentmods.dev/badge/skills/mathaix/openclawmachines/ocm-testing.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 38 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 653 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	$0.00038	$0.00653
Opus 5.5	$0.00015	$0.00261
Sonnet 5.5	$0.00008	$0.00131
Haiku 4.5	$0.00004	$0.00065

Measured 2mo ago against content hash 09ff34751644, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

ocm-testing 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 2mo 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/ocm-testing/SKILL.md · 68 lines

How it starts

The opening of the file, as written. The whole thing — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.

OCM Testing

Use this skill when deciding what to test, debugging failing checks, or validating changes before commit, push, or PR review.

Read First

AGENT.md for repo-wide testing expectations.
Makefile for supported commands.
docs/local-firecracker-e2e.md for local KVM/Firecracker proof.
docs/kvm-integration-ci.md for maintainer-gated KVM runners.
Relevant package docs near the changed code.

Default Rule

Prove the touched surface first. Do not reflexively run every expensive lane.

Inspect the diff and classify touched surfaces.
Reproduce narrowly before fixing when a bug is reported.
Patch the implicated owner module.
Rerun the narrow proof.
Broaden only when a shared contract, integration boundary, or user workflow requires it.

Surface Map

Backend control plane, store, runtime, routing: cd backend && go test ./internal/<pkg>.
Shared backend behavior or broad API changes: make test-go.
Worker routing/auth/CORS/KV behavior: make test-worker.
Frontend UI or auth/client behavior: make test-frontend and make typecheck.
Cross-cutting code changes: make check, make test, make typecheck.
Docs-only: inspect rendered links/commands when practical and run git diff --check.
Scripts/shell automation: make check-scripts if available through make check; otherwise run the script's dry-run/help path when it has one.
Firecracker, rootfs init, agent networking, TAP/bridge/NAT, VM persistence, or KVM placement: run the narrowest KVM integration lane available, or state the exact blocker such as missing /dev/kvm, root privileges, artifacts, network, or Cloudflare credentials.
Cloudflare Worker deployment shape: test Worker logic locally and review worker/wrangler.toml plus docs/self-hosted.env.example for routing loops.

Guardrails

Prefer make targets over direct tool invocations when a target exists.
Do not run privileged KVM, Docker, Cloudflare, GCP, or deploy commands unless the user asked or the change genuinely needs that proof.
Do not kill unrelated processes. Treat running services as user-owned unless they were started in the current task.
Do not print secrets from env files, logs, cookies, JWTs, database URLs, or Cloudflare/Firebase/GCP credentials.
If dependency install or network access is required and blocked, report the blocker and use the closest local proof.

Read the full file on GitHub · 68 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.

2mo ago First seen · 68 lines · 38 tokens per session scan A 09ff34751644

Subscribe to this mod's changes

ocm-testing is a skill published in the GitHub repository mathaix/OpenClawMachines (62 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 38 tokens to every session and 653 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 2026-08-30.

LLM Testing agent-infrastructure ai-agents firecracker kvm

ocm-testing next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
ocm-testingmathaix/OpenClawMachines	Skill	62	3mo ago	Apache-2.0	A	38	Claude Code, Codex
technocore-chatflop-labs/technocore-chat	Skill	167	yesterday	Apache-2.0	A	79	Claude Code, Codex
harness-auditjoymin5655/Agent	Skill	2	yesterday	MIT	A	105	Claude Code
harness-loopjoymin5655/Agent	Skill	2	yesterday	MIT	A	98	Claude Code
code-tourjoymin5655/Agent	Skill	2	yesterday	MIT	A	115	Claude Code
loopjoymin5655/Agent	Skill	2	yesterday	MIT	A	103	Claude Code
harness-helpjoymin5655/Agent	Skill	2	yesterday	MIT	A	41	Claude Code

OpenClawMachines: Skill for Claude Code