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 skills/mathaix/openclawmachines/ocm-integrationsnpx skills add mathaix/OpenClawMachines --skill ocm-integrationsgit clone --depth 1 https://github.com/mathaix/OpenClawMachinesWrote 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.
[](https://agentmods.dev/skills/mathaix/openclawmachines/ocm-integrations)<a href="https://agentmods.dev/skills/mathaix/openclawmachines/ocm-integrations"><img src="https://agentmods.dev/badge/skills/mathaix/openclawmachines/ocm-integrations.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00073 | $0.01150 |
| Opus 5 | $0.00036 | $0.00575 |
| Sonnet 5 | $0.00015 | $0.00230 |
| Haiku 4.5 | $0.00007 | $0.00115 |
Grade A, and why
ocm-integrations 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 5d 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 — 109 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ocm-integrations
OCM workspace integrations are configured at the workspace level and surfaced
inside the machine by the native mcp.servers.ocm MCP server. The backend keeps
upstream credentials server-side; the machine only receives a per-machine token
and policy-filtered OCM facade tools.
Required workflow
- Search first with
ocm.search_toolswhen that tool is available. Usemethod: "semantic"unless the user asks for an exact tool name. Search with the user's intent, not guessed provider function names. - If the provider or access level is known, include
integrationandaccessfilters to keep the result set small. Useaccess: "read"for lookups andaccess: "write"for create, send, update, upload, or delete actions. - Describe exactly one selected result with
ocm.describe_toolbefore calling it. Use thetool_addressreturned by search. - Execute with
ocm.call_toolusing that sametool_addressand arguments that match the described schema.
Prefer the OCM search, describe, and call flow because it keeps tool discovery compact and routes execution through OCM policy, logging, and redaction.
MCP guidance surface
When the MCP client exposes resources or prompts, the same guidance is available
from resource ocm://workspace-integrations/agent-guidance and prompt
ocm_workspace_integrations. Treat those as runtime reminders of this skill;
they do not replace the required search, describe, and call workflow.
Addresses and policy
Search results include legacy tool_id values such as github.search_code and
structured tool_address values such as
wi.<workspace_id>.github.github-main.search_code. Use tool_address because it
distinguishes multiple connections for the same app, such as two Gmail accounts
or two GitHub repos. Use tool_id only when there is no tool_address and the
tool is clearly unambiguous. If a result has policy_state: "require_approval",
report that the tool needs human approval instead of trying to call it.
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.
- 5d ago First seen · 109 lines · 73 tokens per session scan A 68f893893760
ocm-integrations is a skill published in the GitHub repository mathaix/OpenClawMachines (57 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 73 tokens to every session and 1,150 once invoked, about $0.0004 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 skills, from other repositories
kotlin-testing
Kotlin testing patterns with Kotest, MockK, coroutine testing, property-based testing, and Kover coverage. Follows TDD methodology with idiomatic Kotlin practices.
systematic-debugging
4-phase root cause debugging: understand bugs before fixing.
rust-testing
Rust testing patterns including unit tests, integration tests, async testing, property-based testing, mocking, and coverage. Follows TDD methodology.
test-driven-development
TDD: enforce RED-GREEN-REFACTOR, tests before code.
cpp-testing
Use only when writing/updating/fixing C++ tests, configuring GoogleTest/CTest, diagnosing failing or flaky tests, or adding coverage/sanitizers.
tdd
Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.