Security Audit

salesforce-automation

github.com/sickn33/antigravity-awesome-skills

AI SkillCommit e36d6fd3b3f6

TRUSTED

Scanned 1 day ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

salesforce-automation received a trust score of 85/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.

SkillShield's automated analysis identified 1 finding: 0 critical, 1 high, 0 medium, and 0 low severity. Key findings include Unrestricted SOQL Query Tool Allows Data Exfiltration.

The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. All layers scored 70 or above, reflecting consistent security practices.

Last analyzed on February 20, 2026 (commit e36d6fd3). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.

Layer Breakdown

Manifest Analysis

100%

Static Code Analysis

100%

Dependency Graph

100%

LLM Behavioral Safety

85%

Behavioral Risk Signals

Shell Execution

1 finding

Excessive Permissions

1 finding

Security Findings1

Severity	Finding	Layer	Location
HIGH	Unrestricted SOQL Query Tool Allows Data Exfiltration The skill exposes `SALESFORCE_RUN_SOQL_QUERY` and `SALESFORCE_QUERY` tools, which accept an arbitrary `query` string. This allows the LLM (or a malicious user prompt) to execute any SOQL query against the connected Salesforce instance, potentially leading to the exfiltration of sensitive data that the connected user has access to. There are no apparent restrictions or sanitization mechanisms described for the `query` parameter, granting excessive permissions. Implement strict validation and sanitization for the `query` parameter in the `SALESFORCE_RUN_SOQL_QUERY` and `SALESFORCE_QUERY` tools. Ideally, provide more granular tools for common query patterns instead of a raw SOQL interface. If a raw SOQL interface is necessary, ensure it operates with the principle of least privilege and that the LLM is explicitly instructed on how to validate and sanitize user-provided query components. Consider adding an allowlist of accessible objects/fields or a deny-list for sensitive fields.	LLM	SKILL.md:79

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/7a7cb90c9f561d59.svg)](https://skillshield.io/report/7a7cb90c9f561d59)