llm-application-dev-langchain-agent

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill implements an indirect prompt injection surface by directly embedding the $ARGUMENTS variable into the agent's core context.
  • Ingestion points: Untrusted user data enters the system through the $ARGUMENTS placeholder in the 'Context' section of SKILL.md.
  • Boundary markers: Absent. The content is interpolated as plain text without delimiters (such as XML tags or triple quotes) that would help the model distinguish between instructions and data.
  • Capability inventory: The skill instructs the agent to act as a developer capable of writing code, implementing RAG pipelines, and configuring production infrastructure. This high level of agency could be abused if an attacker provides malicious instructions via the $ARGUMENTS variable.
  • Sanitization: Absent. There is no evidence of input validation or filtering before the user data is used to populate the system prompt.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 09:24 PM
Security Audit — agent-trust-hub — llm-application-dev-langchain-agent