AI agents use update_skill to create or update resources in CodeForge MCP — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your CodeForge MCP environment.
The name 'update_skill' implies modifying stored skill data, which is a Write operation. However, given the server context — which includes 'execute_code' (arbitrary code execution) and 'delete_skill' (destructive) — skills likely encapsulate executable code or agent instructions, meaning a malicious update could introduce harmful behavior. This elevates severity to high despite the Write classification.
From the tool's definition Tool name 'update_skill' suggests modifying an existing skill; description is empty and uninformative.
Documented attack patterns abuse exactly the kind of access update_skill gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and CodeForge MCP, and nothing reaches the server without passing your rules. This is the rule we recommend for update_skill:
{
"version": "1",
"default": "deny",
"tools": {
"update_skill": {
"limits": [
{
"counter": "update_skill_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}update_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.
update_skill. It is categorised as a Write tool in the CodeForge MCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the CodeForge MCP server in PolicyLayer and add a rule for update_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 CodeForge MCP. Nothing to install.
update_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 update_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 update_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.
update_skill is provided by the CodeForge MCP server (max-rousseau/codeforge-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from CodeForge MCP, 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.
5 CodeForge MCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.