openexecutive
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes provided Bash and Python scripts (
openexecutive.shandaudit-config.py) to perform system readiness checks and configuration posture audits. These scripts are executed locally using the Bash tool and are designed to be read-only and offline. - [INDIRECT_PROMPT_INJECTION]: The skill manages an application that ingests external data sources such as environment files and company documents for retrieval-augmented generation (RAG). This represents an attack surface for indirect prompt injection. The skill addresses this by providing an offline auditor to validate environment files and including specific instructions to inject retrieval context into the user turn rather than the system prompt to maintain boundary integrity.
- [DATA_EXPOSURE]: The skill interacts with sensitive configuration files (
.env). To prevent accidental exfiltration or logging of secrets, the skill provides a dedicated Python auditor (audit-config.py) that strictly masks all credentials, tokens, and personal identifiers, echoing only a pre-defined allowlist of non-sensitive feature flags and model names.
Audit Metadata