paper_fulltext_search
Search INSIDE the indexed open-access corpus (arXiv + PubMed Central OA full text) for a phrase or keywords and get back the matching passages, each with the paper title, authors, and a snippet around the match. This is the headline feature: agents can find where a finding or method is discussed ...
This record as markdown: /tools/io-github-blackboxfoundry-livedatalink/paper-fulltext-search.md
What paper_fulltext_search does on Livedatalink
AI agents call paper_fulltext_search to retrieve information from Livedatalink without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
| Parameter | Type | Required | Description |
|---|---|---|---|
limit | number | — | Maximum passages to return (default 10, max 50). |
query | string | Yes | Phrase or keywords to find inside the papers, e.g. 'scaled dot-product attention', 'gradient checkpointing'. |
paper_key | string | — | Optional: restrict the search to a single indexed paper by its corpus key, e.g. 'arxiv:2310.12345'. |
Parameters from the server's own tool schema.
Why paper_fulltext_search is rated Low
This is a standard information retrieval tool that searches and returns data from public academic repositories (arXiv, PubMed Central). It has no side effects, does not execute code or commands, does not modify or delete data, and does not involve financial transactions. The primary action is querying/searching an indexed corpus to retrieve matching text passages, which is characteristic of Read category tools.
From the tool's definition Tool performs search/query operations ("Search INSIDE the indexed open-access corpus", "find where a finding or method is discussed") and returns read-only results ("matching passages", "paper title, authors, and a snippet").
Attacks that exploit this kind of access
The rule that runs paper_fulltext_search safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Livedatalink, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For paper_fulltext_search, this is the rule to start with:
paper_fulltext_search is read-only, so it stays allowed. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Livedatalink, apply this rule, and every paper_fulltext_search call is checked against it from then on.
Questions about paper_fulltext_search
Search INSIDE the indexed open-access corpus (arXiv + PubMed Central OA full text) for a phrase or keywords and get back the matching passages, each with the paper title, authors, and a snippet around the match. This is the headline feature: agents can find where a finding or method is discussed across open-access papers. Optionally restrict to one paper by paper_key. It is categorised as a Read tool in the Livedatalink MCP Server, which means it retrieves data without modifying state.
paper_fulltext_search accepts 3 parameters: limit, query, paper_key. Required: query. The full parameter table on this page comes from the server's own tool schema.
Register the Livedatalink MCP server in PolicyLayer and add a rule for paper_fulltext_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 Livedatalink. Nothing to install.
paper_fulltext_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 paper_fulltext_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 paper_fulltext_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.
paper_fulltext_search is provided by the Livedatalink MCP server (https://livedatalink.ai/mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Livedatalink, and thousands of servers like it.
This server
Across the catalogue