create-generation-plugin

Pass

Audited by Gen Agent Trust Hub on Jun 20, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: No malicious prompt injection patterns, safety bypasses, or instruction overrides were detected in the instructions.
  • [DATA_EXFILTRATION]: No evidence of unauthorized data access, hardcoded credentials, or network exfiltration patterns. All operations are local to the development environment.
  • [REMOTE_CODE_EXECUTION]: The skill uses local commands such as pip install -e . to install a locally developed package, which is standard practice for Python development. No remote scripts are downloaded or executed from external sources.
  • [COMMAND_EXECUTION]: Shell commands are restricted to local project scaffolding (autorag-research plugin create) and environment management, which align with the skill's stated purpose as a developer guide.
  • [INDIRECT_PROMPT_INJECTION]: While the skill guides the creation of RAG pipelines which ingest external data, this represents a standard functional architecture for the framework and does not introduce malicious surfaces beyond the inherent nature of RAG systems.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 20, 2026, 06:08 AM
Security Audit — agent-trust-hub — create-generation-plugin