AI agents use cache_export to create or update resources in Crossref Local — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Crossref Local environment.
This tool creates or modifies files by exporting cached data. It is reversible (exported files can be deleted or replaced), making it Write rather than Destructive. The severity is medium because exporting data itself has limited blast radius—the risk is data exposure or file system clutter rather than data loss or external impact.
From the tool's definition Tool name is 'cache_export' and description states 'Export cache to file.' The verb 'export' combined with 'to file' indicates the tool writes data to persistent storage.
Documented attack patterns abuse exactly the kind of access cache_export gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Crossref Local, and nothing reaches the server without passing your rules. This is the rule we recommend for cache_export:
{
"version": "1",
"default": "deny",
"tools": {
"cache_export": {
"limits": [
{
"counter": "cache_export_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}cache_export 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.
Export cache to file. It is categorised as a Write tool in the Crossref Local MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Crossref Local MCP server in PolicyLayer and add a rule for cache_export: 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 Crossref Local. Nothing to install.
cache_export 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 cache_export 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 cache_export. 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.
cache_export is provided by the Crossref Local MCP server (ywatanabe1989/crossref-local). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Crossref Local, 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.
17 Crossref Local tools catalogued and risk-classified — across an index of 43,000+ MCP servers.