Change logging config at runtime
AI agents use kaiten_set_log_level to create or update resources in Kaiten — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Kaiten environment.
This tool modifies logging configuration, which is a reversible write operation affecting system behavior rather than core data. It does not delete data (Destructive), execute external code (Execute), create business logic (Write to data), or involve financial transactions (Financial).
From the tool's definition Tool name 'kaiten_set_log_level' and description 'Change logging config at runtime' indicate modification of logging configuration settings.
Documented attack patterns abuse exactly the kind of access kaiten_set_log_level gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Kaiten, and nothing reaches the server without passing your rules. This is the rule we recommend for kaiten_set_log_level:
{
"version": "1",
"default": "deny",
"tools": {
"kaiten_set_log_level": {
"limits": [
{
"counter": "kaiten_set_log_level_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}kaiten_set_log_level 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.
Change logging config at runtime. It is categorised as a Write tool in the Kaiten MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Kaiten MCP server in PolicyLayer and add a rule for kaiten_set_log_level: 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 Kaiten. Nothing to install.
kaiten_set_log_level 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 kaiten_set_log_level 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 kaiten_set_log_level. 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.
kaiten_set_log_level is provided by the Kaiten MCP server (vsaranyuk/kaiten-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Kaiten, 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.
8 Kaiten tools catalogued and risk-classified — across an index of 43,000+ MCP servers.