llm-application-dev-langchain-agent
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides legitimate software development guidance and code snippets for LangChain and LangGraph development.
- [INDIRECT_PROMPT_INJECTION]: The skill incorporates user-provided requirements through the
$ARGUMENTSvariable. - Ingestion points: The Context section in SKILL.md accepts external input via
$ARGUMENTS. - Boundary markers: No explicit delimiters are used around the interpolated arguments.
- Capability inventory: The skill is designed for code generation and architectural consultation.
- Sanitization: No sanitization is performed on the input, but the risk is inherently low as the input is used to scope a technical discussion.
Audit Metadata