AI agents use save_pipeline to create or update resources in Pubmed Search — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Pubmed Search environment.
The 'save' operation typically creates or modifies data (Write category). In a research assistant context, this likely persists user-defined pipeline configurations. While potentially reversible via deletion (unlike Destructive), the tool modifies persistent state.
From the tool's definition Tool name 'save_pipeline' indicates creation or modification of a pipeline configuration. Context from sibling tool 'configure_institutional_access' and 'delete_pipeline' suggests state management of research analysis pipelines.
Documented attack patterns abuse exactly the kind of access save_pipeline gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Pubmed Search, and nothing reaches the server without passing your rules. This is the rule we recommend for save_pipeline:
{
"version": "1",
"default": "deny",
"tools": {
"save_pipeline": {
"limits": [
{
"counter": "save_pipeline_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}save_pipeline 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.
save_pipeline. It is categorised as a Write tool in the Pubmed Search MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Pubmed Search MCP server in PolicyLayer and add a rule for save_pipeline: 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 Pubmed Search. Nothing to install.
save_pipeline 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 save_pipeline 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 save_pipeline. 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.
save_pipeline is provided by the Pubmed Search MCP server (u9401066/pubmed-search-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Pubmed Search, 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.
63 Pubmed Search tools catalogued and risk-classified — across an index of 43,000+ MCP servers.