service-de-headless-channel-configure
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the Salesforce CLI (sf) to interact with the target organization, executing commands such as 'sf data query' to detect current state and perform configuration. This usage is appropriate for a DevOps/administrative orchestration skill provided by the vendor.
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect injection surface by interpolating user-provided inputs like '{PLATFORM_KEY}' and '{MESSAGE_TYPE}' directly into SOQL query strings.
- Ingestion points: User inputs for channel IDs and types are ingested in Stage 0 and used in Stage 1 (SKILL.md).
- Boundary markers: No explicit delimiters or escaping mechanisms are applied to these variables before they are included in the shell-executed query string.
- Capability inventory: The skill has the capability to execute database queries ('sf data query') and modify organization settings ('sf data patch', etc., via leaf skills).
- Sanitization: The skill lacks explicit sanitization or validation of the input fields, which could potentially allow for query injection if provided with crafted payloads.
- [SAFE]: A mandatory blocking gate is implemented in Stage 0.5 (SKILL.md), which renders a legal disclaimer and requires the user to certify their authority before proceeding. This ensures that the autonomous execution of subsequent configuration stages is preceded by informed user consent, mitigating risks associated with automated provisioning.
Audit Metadata