Unverified50% confidenceFactExact time
In the LLM tool-calling paradigm, the model only outputs structured parameters, which the host program validates against a predefined JSON Schema and binds to precompiled Cypher query templates, so user or alert text content is only ever passed in as parameter values, making it naturally immune to Cypher injection
1
Sources
50%
Confidence
Long-term
Relevance
10/18/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedCurrent mainstream LLM agents do not execute code or SQL directly; instead, the model outputs structured text (usually JSON) declaring the functions and parameters to be called, which the host program then parses and executes.76% similarVerifiedUnlike rule-based static analysis tools (such as CodeQL) or symbolic execution tools, LLMs can understand comment intent, recognize the semantics of naming conventions, and infer data flow in the absence of complete type information.69% similarUnverifiedLLM tool calling (Tool Calling/Function Calling) is a capability of large models: the model generates structured tool-call requests according to a preset JSON Schema, external code executes them and passes the results back to the model, and the model continues reasoning based on the results.68% similarVerifiedThe basic principle of tool calling is that the model identifies scenarios requiring external capabilities, outputs a structured function call request (usually in JSON format), and after the external system executes it, the result is returned to the model to continue reasoning.68% similarUnverifiedIn LLM agent scenarios, the model treats tool description fields as part of its own configuration to follow and treats result fields as external data to process, and this difference in trust level causes the same text to carry different injection risk in different fields.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1062916API
curl https://kongchang.com/api/v1/knowledge/claims/1062916MCP
get_claim(id=1062916)