salesforce-cost-tuning
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external data from Salesforce, which represents a potential surface for indirect prompt injection if the processed data contains malicious instructions.
- Ingestion points: The instructions in
SKILL.mddescribe processing data from the Salesforce REST API/limits/endpoint, record collections (contacts), and external CSV files (csvData). - Boundary markers: The provided code snippets do not demonstrate the use of delimiters or explicit instructions for the agent to ignore instructions embedded within the ingested data.
- Capability inventory: The skill logic involves network-based API requests and data modification operations (create, bulk load) within the Salesforce environment.
- Sanitization: No specific data validation or sanitization routines are shown for the record data or CSV inputs before they are passed to the Salesforce API.
Audit Metadata