dx-org-manage
Pass
Audited by Gen Agent Trust Hub on Nov 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements strong credential protection by requiring the agent to redact sensitive Salesforce fields, including 'accessToken' and 'sfdxAuthUrl', from all command results using 'jq' filters before they are stored as artifacts or processed further.\n- [SAFE]: Security-focused constraints are applied to CLI tools, specifically banning the '--verbose' and '--url-only' flags for org display and open operations to prevent session-equivalent tokens from entering the agent's context. The skill correctly instructs users to run these commands manually if the raw tokens are needed.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from local project configuration files and external CLI tool outputs, representing a potential injection surface. \n
- Ingestion points: Reads 'sfdx-project.json', various scratch org definition files in the 'config/' directory, and structured JSON output from the 'sf' CLI. \n
- Boundary markers: No explicit delimiters or 'ignore' instructions are used for the content parsed from these files. \n
- Capability inventory: The skill possesses the capability to execute shell commands via the Bash tool and write JSON response files to the local file system. \n
- Sanitization: Employs 'jq' to explicitly delete credential fields from ingested data before it is utilized in subsequent steps or saved to disk.
Audit Metadata