GCD AI RADAR
incidents/AIR-2023-0005

LLM Application Frameworks Allow Remote Code Execution via Prompting

Attack on AI1 Research / proof of conceptPrompt injectionAgent hijack

Frameworks like langchain and boxcars.ai allow direct query execution through LLMs, enabling remote code execution or SQL injection via prompt inputs. Researchers demonstrated that unauthorized actions could be triggered through conversational input.

Details

type
Attack on AI
An AI model, agent or AI-powered product was the thing exploited, abused or leaked.
categories
  • Prompt injectionprimary
    Untrusted content (web pages, documents, emails, tool output) steers a model into ignoring its instructions.
  • Agent hijack
    An AI agent is tricked into unauthorized actions: tool misuse, data exfiltration through the agent, or acting against its principal.
owasp llm top 10
mitre atlas
Both rows are derived from the type and categories (OWASP 2025 edition, ATLAS release 2026.09), not classified per record. How
severity
1 · Research / proof of concept
Demonstrated by researchers; no known real-world exploitation.
targets
OpenAI, Framework / tooling
products
ChatGPT, langchain, boxcars.ai
incident date
26 Mar 2023
first reported
26 Mar 2023
origin
AVID
classifier confidence
0.90 · published at or above 0.80; see the methodology
tags
prompt-injection · remote-code-execution · langchain
record
created 2026-10-18 06:56Z · updated 2026-10-18 06:56Z · json

Sources · 4

  1. 2023-03-26
    AVID-2023-R0003: Frameworks like langchain (Python) and boxcars.ai (Ruby) offer apps and scripts to directly execute queries through LLMs as a built-in feature.
    AVID · github.com/avidml/avid-db/blob/main/reports/2023/AVID-2023-R0003.json
  2. 2023-03-26
    InjectGPT: the most polite exploit ever
    blog.luitjes.it · blog.luitjes.it/posts/injectgpt-most-polite-exploit-ever
  3. 2023-03-26
    Reddit thread on InjectGPT
    reddit.com · reddit.com/r/netsec/comments/121gpay/injectgpt_remote_code_execution_by_asking_n
  4. 2023-03-31
    AVID-2023-V028: Frameworks like langchain (Python) and boxcars.ai (Ruby) offer apps and scripts to directly execute queries through LLMs as a built-in feature.
    AVID · github.com/avidml/avid-db/blob/main/vulnerabilities/2023/AVID-2023-V028.json

Titles and links belong to their publishers. The summary above is written by GCD AI RADAR in its own words; data is CC BY-SA 4.0.

Related incidents