AI agents invoke beancount_query to trigger actions in Beancount. 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.
The description is empty, so classification relies on the tool name and server context. 'beancount_query' most likely executes BQL (Beancount Query Language) queries against a financial ledger.
From the tool's definition Tool name is 'beancount_query' on a server described as allowing users to 'query from your Beancount ledger, or submit transaction.' Sibling tools include transaction submission and replacement.
Documented attack patterns abuse exactly the kind of access beancount_query gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Beancount, and nothing reaches the server without passing your rules. This is the rule we recommend for beancount_query:
{
"version": "1",
"default": "deny",
"tools": {
"beancount_query": {
"limits": [
{
"counter": "beancount_query_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}beancount_query 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.
beancount_query. It is categorised as a Execute tool in the Beancount MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Beancount MCP server in PolicyLayer and add a rule for beancount_query: 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 Beancount. Nothing to install.
beancount_query 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 beancount_query 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 beancount_query. 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.
beancount_query is provided by the Beancount MCP server (stdioa/beancount-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Beancount, 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 Beancount tools catalogued and risk-classified — across an index of 43,000+ MCP servers.