AI agents use create_new_skill to create or update resources in Skill Cortex Lite — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Skill Cortex Lite environment.
Creating new skills in a code skills system is a reversible write operation. While the tool description is empty (lowering confidence slightly), the name clearly indicates data creation rather than deletion or execution. The medium severity reflects that skill creation could introduce malicious or unwanted code skills into the IDE environment, but the effect is reversible via the sibling 'delete_existing_skill' tool.
From the tool's definition Tool name 'create_new_skill' indicates creation of new data objects (skills). The server context shows this is part of a skill management system for IDEs.
Documented attack patterns abuse exactly the kind of access create_new_skill gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Skill Cortex Lite, and nothing reaches the server without passing your rules. This is the rule we recommend for create_new_skill:
{
"version": "1",
"default": "deny",
"tools": {
"create_new_skill": {
"limits": [
{
"counter": "create_new_skill_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}create_new_skill 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.
Free to start. No card required.
create_new_skill. It is categorised as a Write tool in the Skill Cortex Lite MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Skill Cortex Lite MCP server in PolicyLayer and add a rule for create_new_skill: 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 Skill Cortex Lite. Nothing to install.
create_new_skill is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the create_new_skill 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.
Set action: deny in the PolicyLayer policy for create_new_skill. 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.
create_new_skill is provided by the Skill Cortex Lite MCP server (sim-xia/skill-cortex-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Skill Cortex Lite, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
6 Skill Cortex Lite tools catalogued and risk-classified — across an index of 43,000+ MCP servers.