Unverified50% confidenceConceptExact time
Tool use (Tool Use / Function Calling) lets LLMs perform real actions, escalating the consequences of prompt injection from outputting incorrect text to triggering real side effects.
1
Sources
50%
Confidence
Long-term
Relevance
10/19/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedTool Calling allows an LLM to decide, during response generation, to invoke predefined external functions or APIs and incorporate the results into subsequent reasoning.75% similarPartially VerifiedPrompt injection is a class of attack faced by LLM applications in which attackers disguise malicious instructions as normal content to induce the model to execute them.70% similarUnverifiedThe assistant message is used to simulate the LLM's text reply, and the function call is used to simulate the invocation instruction generated when the LLM decides to call a tool70% similarUnverifiedIn systems with tools attached, if the model can call APIs, send messages, write to databases, or execute code, injected instructions that get executed may cause irreversible consequences.68% similarUnverifiedThe Function Calling mechanism upgrades LLMs from being able to only generate text to being able to operate external systems, and is the technical foundation for agents in Agentic CRM to perform actions such as data writing and status updates.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1071259API
curl https://kongchang.com/api/v1/knowledge/claims/1071259MCP
get_claim(id=1071259)