hybrid-retrieval

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE] (SAFE): The skill consists of educational documentation and Python code snippets for semantic and keyword search fusion.
  • [Category 1: Prompt Injection]: No malicious instructions or overrides were found in the skill body or metadata.
  • [Category 2: Data Exposure]: No sensitive file access or hardcoded credentials detected. Placeholder keys (e.g., 'api_key="..."') are used appropriately for demonstration.
  • [Category 3: Obfuscation]: No encoded commands, zero-width characters, or homoglyphs are present.
  • [Category 4 & 10: Code Execution]: The code snippets use reputable libraries (LangChain, Pinecone, Weaviate) for their intended purpose. There is no evidence of remote script execution or unsafe dynamic execution.
  • [Category 8: Indirect Prompt Injection]: While the skill defines a retrieval surface that processes user queries, it does not implement a full agent loop or execute instructions found in retrieved content, and it lacks the exploitable capabilities required for this risk to manifest within the skill's scope.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:40 PM