langchain-sdk-patterns
Pass
Audited by Gen Agent Trust Hub on Mar 24, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect prompt injection surface identified in the instructional code examples.
- Ingestion points: The skill in
SKILL.mddemonstrates patterns using{text}variables within prompt templates, such asAnalyze the sentiment of: {text}andSummarize: {text}. - Boundary markers: The provided examples do not use delimiters or explicit instructions to the model to ignore embedded commands within the input data.
- Capability inventory: The patterns demonstrate LLM invocation via
invoke,abatch, andstream, which can be influenced by untrusted data if the agent executes this code with external inputs. - Sanitization: No input validation, filtering, or escaping logic is included in the demonstrated SDK patterns.
Audit Metadata