Trust Assessment
voice-call received a trust score of 93/100, placing it in the Trusted category. This skill has passed all critical security checks and demonstrates strong security practices.
SkillShield's automated analysis identified 2 findings: 0 critical, 0 high, 1 medium, and 0 low severity. Key findings include Broad Tool Access: Arbitrary Call Initiation, Sensitive Credentials in Configuration.
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 18, 2026 (commit b62bd290). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings2
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| MEDIUM | Broad Tool Access: Arbitrary Call Initiation The `initiate_call` action allows the AI agent to make voice calls to arbitrary phone numbers (`to?`) with arbitrary messages. This capability, if misused or exploited by a compromised LLM, could lead to unwanted calls, spam, social engineering attempts, or other forms of communication abuse. While the skill's intended purpose is to make calls, the lack of explicit constraints on the 'to' parameter within the tool description itself represents a broad permission that requires careful management by the orchestrating LLM and user guardrails. Implement strong guardrails and user consent mechanisms in the orchestrating LLM to prevent unauthorized or malicious use of the `initiate_call` function. Consider adding configuration options to restrict callable numbers (e.g., allow-lists, block-lists) or require explicit user confirmation for calls to unapproved numbers. Monitor usage for suspicious patterns. | Static | SKILL.md:20 | |
| INFO | Sensitive Credentials in Configuration The skill's documentation indicates that highly sensitive credentials such as Twilio `accountSid/authToken`, Telnyx `apiKey/connectionId`, and Plivo `authId/authToken` are stored directly within the plugin's configuration (`plugins.entries.voice-call.config`). While this document does not exfiltrate data, it highlights that these critical secrets will reside in the system's configuration. Improper protection of this configuration file could lead to credential compromise. Ensure that the configuration file containing these credentials is secured with appropriate file system permissions, encryption at rest, and access controls. Avoid hardcoding credentials directly in code or committing them to version control. Utilize secure secret management solutions (e.g., environment variables, dedicated secret stores) where possible, and rotate credentials regularly. | Static | SKILL.md:26 |
Scan History
Embed Code
[](https://skillshield.io/report/5bec64cb5fbcc299)
Powered by SkillShield