skill_create

Create a new custom skill. The skill will be saved to the managed skills directory

SERVERCoWork OS SOURCECoWork-OS/CoWork-OS

Medium RISK CLASS

Category Write

Parameters 00 required

Recommended Rate-limitedsee the rule below

Registry record Grade F, identity unverified Pull the record →

This record as markdown: /tools/cowork-os/skill-create.md

WHAT IT DOES

What skill_create does on CoWork OS

AI agents use skill_create to create or update resources in CoWork OS, usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your CoWork OS environment.

RISK

Why skill_create is rated Medium

This tool creates and persistently stores new custom skills, which is a reversible write operation. It does not execute arbitrary code immediately (Execute), nor does it delete data (Destructive). The severity is medium because creating malicious or unintended skills could affect system behavior, but the effect is localized to skill definitions and can be modified or removed.

From the tool's definition Tool description states 'Create a new custom skill' and 'will be saved to the managed skills directory', indicating creation and storage of new data/artifacts.

Attacks that exploit this kind of access

RECOMMENDED RULE

The rule that runs skill_create safely

PolicyLayer is an MCP gateway: it sits between your AI agents and CoWork OS, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For skill_create, this is the rule to start with:

skill_create Rate-limited

skill_create stays usable, but capped: an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "skill_create": {
      "limits": [
        {
          "counter": "skill_create_rate",
          "window": "minute",
          "max": 30,
          "scope": "grant"
        }
      ]
    }
  }
}

LIMIT THIS TOOL → Instant setup, no code required.

The button opens the PolicyLayer dashboard: create your workspace, connect CoWork OS, apply this rule, and every skill_create call is checked against it from then on.

FAQ

Questions about skill_create

What does the skill_create tool do? +

Create a new custom skill. The skill will be saved to the managed skills directory. It is categorised as a Write tool in the CoWork OS MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

How do I enforce a policy on skill_create? +

Register the CoWork OS MCP server in PolicyLayer and add a rule for skill_create: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches CoWork OS. Nothing to install.

What risk level is skill_create? +

skill_create is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.

Can I rate-limit skill_create? +

Yes. Add a rate_limit block to the skill_create rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.

How do I block skill_create completely? +

Set action: deny in the PolicyLayer policy for skill_create. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.

What MCP server provides skill_create? +

skill_create is provided by the CoWork OS MCP server (CoWork-OS/CoWork-OS). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

KEEP EXPLORING

More on CoWork OS, and thousands of servers like it.

This server

All 342 CoWork OS tools →

Across the catalogue

create_task on 0rca Dojo MCP Server →create-task on Kanban MCP Server →create_package_backup on Android Forensics ADB MCP Server →create_project on Stitch MCP Server → The MCP Attack Database →

Guides

Govern customer-ops agents →Argument validation →Scoped token →