AI agents invoke kobold_chat to trigger actions in Kobold MCP Server. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool executes a chat completion request against KoboldAI's text generation engine. It initiates an external AI inference operation whose output depends on the input arguments. While it doesn't modify stored data or move money, it actively triggers computation on an external service, placing it in the Execute category. Severity is medium as misuse could generate harmful content or consume compute resources.
From the tool's definition Chat completion (OpenAI-compatible) — triggers text generation via KoboldAI backend
Documented attack patterns abuse exactly the kind of access kobold_chat gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Kobold MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for kobold_chat:
{
"version": "1",
"default": "deny",
"tools": {
"kobold_chat": {
"limits": [
{
"counter": "kobold_chat_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}kobold_chat stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Chat completion (OpenAI-compatible). It is categorised as a Execute tool in the Kobold MCP Server MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Kobold MCP Server MCP server in PolicyLayer and add a rule for kobold_chat: 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 Kobold MCP Server. Nothing to install.
kobold_chat is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the kobold_chat 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 kobold_chat. 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.
kobold_chat is provided by the Kobold MCP Server MCP server (phialsbasement/koboldcpp-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Kobold MCP Server, 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.
20 Kobold MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.