embedding-pipelines
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes external data sources which creates a surface for indirect prompt injection.
- Ingestion points: Untrusted data enters the agent context through configuration files like
pipeline.yamland datasets such asdata.csvortext_data.csv. - Boundary markers: The skill instructions do not specify any delimiters or safety warnings to ensure the agent ignores instructions potentially embedded within the ingested data.
- Capability inventory: The skill has the capability to execute CLI commands (
openclaw) and perform network operations via a Python SDK (openclaw-sdk) to deploy and optimize pipelines. - Sanitization: There is no indication that the contents of the ingested files are sanitized or validated against a strict schema before being processed by the agent.
Audit Metadata