salesforce-developer
Fail
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: CRITICALDATA_EXFILTRATIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: Automated scanners have detected a malicious documentation URL (https://jeffallan.github.io/claude-skills/skills/platform/salesforce-developer/) and flagged the main skill file (SKILL.md) for malicious reputation.
- [DYNAMIC_EXECUTION]: The skill provides reference guides and code examples for dynamic SOQL and SOSL queries (Database.query and Search.query) as well as programmatic metadata deployment. While the examples include sanitization best practices, dynamic execution patterns increase the potential for injection vulnerabilities if misused.
- [EXTERNAL_DOWNLOADS]: The deployment documentation (references/deployment-devops.md) includes instructions to download and execute the Salesforce CLI tool from a well-known vendor domain (developer.salesforce.com).
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted business requirements and existing source code to generate implementation artifacts, which can be misused via indirect prompt injection to execute unauthorized environment commands.
- Ingestion points: Business requirement analysis and existing code debugging (SKILL.md).
- Boundary markers: None. The skill does not provide instructions to ignore malicious directives within analyzed requirements or code.
- Capability inventory: Execution of Salesforce CLI commands for environment authentication, metadata deployment, and data management.
- Sanitization: Absent for processing business requirements.
Recommendations
- CRITICAL: 1 file(s) identified as malware by FileRep - DO NOT USE
- AI detected serious security threats
- Contains 2 malicious URL(s) - DO NOT USE
Audit Metadata