RAG Pipeline Builder
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- Prompt Injection (SAFE): No instructions attempting to override agent behavior, bypass safety filters, or extract system prompts were detected.
- Data Exposure & Exfiltration (SAFE): There are no hardcoded credentials, sensitive file path accesses, or unauthorized network operations present in the skill content.
- Remote Code Execution (SAFE): The Python code provided consists of static logic templates for chunking and query processing. The skill does not include commands to download or execute external scripts (e.g., curl/wget piped to bash).
- Obfuscation (SAFE): All content is provided in clear markdown and standard code snippets with no evidence of encoding, zero-width characters, or hidden payloads.
- Indirect Prompt Injection (LOW): The skill is designed to build systems that ingest untrusted external data (PDFs, HTML, APIs). Evidence Chain: 1. Ingestion points: Document loaders and API connectors in Workflow 2. 2. Boundary markers: Best practices explicitly recommend citations and showing source chunks to the user. 3. Capability inventory: No script execution or file-write capabilities are provided within the skill's own code. 4. Sanitization: Recommends handling edge cases where retrieval fails or information is conflicting.
Audit Metadata