AI agents use log_decision to create or update resources in Memory Bank MCP — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Memory Bank MCP environment.
The tool creates or appends a new record (a decision log entry) to persistent storage. This is a reversible write operation — the log entry can typically be modified or removed later. It does not delete data (Destructive), execute arbitrary code (Execute), involve financial transactions (Financial), or merely read data (Read).
From the tool's definition Tool name is 'log_decision' with description 'Log a decision in the decision log' — creates or records a new entry in a decision log, which is a form of data creation/modification.
Documented attack patterns abuse exactly the kind of access log_decision gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Memory Bank MCP, and nothing reaches the server without passing your rules. This is the rule we recommend for log_decision:
{
"version": "1",
"default": "deny",
"tools": {
"log_decision": {
"limits": [
{
"counter": "log_decision_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}log_decision 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.
Log a decision in the decision log. It is categorised as a Write tool in the Memory Bank MCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Memory Bank MCP server in PolicyLayer and add a rule for log_decision: 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 Memory Bank MCP. Nothing to install.
log_decision 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 log_decision 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 log_decision. 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.
log_decision is provided by the Memory Bank MCP server (movibe/memory-bank-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Memory Bank 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.
15 Memory Bank MCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.