What is ocm-debugging?

A debugging guide for OpenClaw Machines, a system that runs agents through hosted services, worker routing, hosts, APIs, and virtual machines. It helps locate which boundary is responsible when behavior differs between environments.

Use it to investigate configuration and authentication, Cloudflare Worker routing, host enrollment, placement, backend APIs, guest startup, Firecracker networking or storage, and runtime version selection.

It replaces guesswork with a repeatable process: identify the suspected layer, add a focused signal, reproduce the same conditions, compare configured and active state, and verify the fix.

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

Your own site

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

Per session 44 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 583 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	$0.00044	$0.00583
Opus 5.5	$0.00018	$0.00233
Sonnet 5.5	$0.00009	$0.00117
Haiku 4.5	$0.00004	$0.00058

Measured 2mo ago against content hash d9f4bd34ee20, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

ocm-debugging 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-debugging/SKILL.md · 71 lines

How it starts

The opening of the file, as written. The whole thing — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.

OCM Debugging

Use this skill when behavior differs between local, operator, hosted-like, Worker, agent, guest VM, or KVM integration paths and the next move should be a debug signal rather than a guess.

Default Loop

State the suspected boundary:
- profile/config
- auth/cookie/session exchange
- Cloudflare Worker/KV routing
- host enrollment or heartbeat
- placement/capacity
- backend-to-agent API
- metadata server
- guest init/rootfs
- Firecracker network/storage
- runtime artifact/version selection
Add or enable the narrowest signal that proves that boundary.
Reproduce with the same profile, env, domain shape, host state, and runtime version. Do not switch profiles or providers unless that variable is under test.
Compare configured state with actual activated state.
Patch the root cause.
Rerun the exact failing probe, then broaden only if a shared contract needs it.

Useful Starting Points

Config/profile load: backend/internal/config.
Session cookies and auth: backend/internal/api/auth_handlers.go.
Worker route selection and resolve fallback: worker/worker.js.
Route provisioning and signing keys: backend/internal/routing.
Machine start/upgrade lifecycle: backend/internal/machines/runtime.go.
Host enrollment: backend/internal/api/enrollment.go.
Agent control API: backend/internal/agentapi.
Firecracker orchestration: backend/internal/orchestrator.
Guest metadata: backend/internal/metadata.
Guest init: scripts/init-openclaw.sh.
Local full-stack runbook: docs/local-firecracker-e2e.md.

Proof Choices

Single backend invariant: focused Go package test.
Worker route/auth bug: make test-worker.
Frontend/auth UI bug: focused frontend test plus make typecheck.
Local control-plane setup: make local-status or the relevant scripts/local-dev.sh command.
Firecracker/agent/rootfs/network bug: KVM integration or local scripts/local-e2e-firecracker.sh status, with exact blocker reported when unavailable.
Cloudflare/operator routing bug: local Worker test plus config/doc review; live Cloudflare proof only when credentials and operator approval are present.

Read the full file on GitHub · 71 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 · 71 lines · 44 tokens per session scan A d9f4bd34ee20

Subscribe to this mod's changes

ocm-debugging is a skill published in the GitHub repository mathaix/OpenClawMachines (62 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 44 tokens to every session and 583 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.

Debugging LLM agent-infrastructure ai-agents firecracker kvm

ocm-debugging next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
ocm-debuggingmathaix/OpenClawMachines	Skill	62	3mo ago	Apache-2.0	A	44	Claude Code, Codex
systematic-debuggingNousResearch/hermes-agent	Skill	261k	yesterday	MIT	A	16	Claude Code, Codex
python-debugpyNousResearch/hermes-agent	Skill	261k	yesterday	MIT	B	17	Claude Code, Codex
sglang-diffusion-benchmark-profilesgl-project/sglang	Skill	39k	yesterday	Apache-2.0	A	26	Claude Code, Codex
resolve-handoffomnigent-ai/omnigent	Skill	11k	yesterday	Apache-2.0	A	23	Claude Code, Codex
iterate-implementationhumanlayer/humanlayer	Skill	13k	yesterday	Apache-2.0	A	16	Claude Code, Codex
verify-omnigentomnigent-ai/omnigent	Skill	11k	yesterday	Apache-2.0	A	100	Claude Code, Codex

OpenClawMachines: Skill for Claude Code