Real-time federated knowledge search (research_search v2 — replaces the retired universal_knowledge raw-dump). Fans out across 10 curated public APIs (PubMed, Europe PMC, ClinicalTrials, OpenFDA, OpenAlex, Crossref, arXiv, Semantic Scholar, Wikipedia, Wikidata), deduplicates cross-source by DOI/t...
AI agents call research_search to retrieve information from Celiums Memory without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
research_search is a retrieval-only tool that queries external public knowledge sources (PubMed, arXiv, Wikipedia, etc.) and returns ranked results. It has no capability to modify, delete, execute code, or trigger external actions beyond read-only API calls. The worst-case misuse would be retrieving sensitive or proprietary information already publicly indexed, which poses minimal risk.
From the tool's definition Tool description states it "fans out across 10 curated public APIs" to perform "federated knowledge search" and "returns ranked results with name, display_name, description, category (source), relevance score, plus authors/year/doi/url/consensus." The verb…
Documented attack patterns abuse exactly the kind of access research_search gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Celiums Memory, and nothing reaches the server without passing your rules. This is the rule we recommend for research_search:
{
"version": "1",
"default": "deny",
"tools": {
"research_search": {}
}
}research_search is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Real-time federated knowledge search (research_search v2 — replaces the retired universal_knowledge raw-dump). Fans out across 10 curated public APIs (PubMed, Europe PMC, ClinicalTrials, OpenFDA, OpenAlex, Crossref, arXiv, Semantic Scholar, Wikipedia, Wikidata), deduplicates cross-source by DOI/title, and fuses with Reciprocal Rank Fusion so multi-source consensus ranks highest. A query-domain router selects the relevant APIs automatically. Returns ranked results with name, display_name, description, category (source), relevance score, plus authors/year/doi/url/consensus. Use to locate evidence before synthesize. It is categorised as a Read tool in the Celiums Memory MCP Server, which means it retrieves data without modifying state.
Register the Celiums Memory MCP server in PolicyLayer and add a rule for research_search: 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 Celiums Memory. Nothing to install.
research_search is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the research_search 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 research_search. 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.
research_search is provided by the Celiums Memory MCP server (terrizoaguimor/celiums-memory). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Celiums Memory, 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.
62 Celiums Memory tools catalogued and risk-classified — across an index of 43,000+ MCP servers.