Skills
skills/aradotso/ai-agent-skills/production-grade-agentic-system/Gen Agent Trust Hub

production-grade-agentic-system

Warn

Audited by Gen Agent Trust Hub on May 17, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill's installation instructions require cloning a codebase from an external GitHub repository (FareedKhan-dev/production-grade-agentic-system).
  • [REMOTE_CODE_EXECUTION]: The setup process involves running pip install -e . on the cloned external repository, which executes arbitrary code contained within the project's installation and setup files.
  • [PROMPT_INJECTION]: The system architecture design presents a surface for indirect prompt injection by ingesting untrusted user data into the agent's context.
  • Ingestion points: The message field from the ChatRequest schema in the chat_endpoint and chat_stream functions.
  • Boundary markers: Absent; there are no delimiters or instructions provided to the LLM to ignore potentially malicious content within the user messages.
  • Capability inventory: The system includes capabilities such as web searching via DuckDuckGoSearchRun and persistent database storage via PostgresSaver and SQLModel.
  • Sanitization: Absent; no validation, escaping, or filtering of the user-provided message is implemented before it is processed by the LangGraph nodes.
Audit Metadata
Risk Level
MEDIUM
Analyzed
May 17, 2026, 10:23 PM
Security Audit — agent-trust-hub — production-grade-agentic-system