connector-twilio
Fail
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: CRITICALCREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the twilio-client package via the command mops add twilio-client@0.1.2. The primary repository URL https://mops.one/twilio-client has a confirmed malicious detection for phishing from automated scanners.
- [REMOTE_CODE_EXECUTION]: The instruction for the agent to install external packages from a flagged malicious source allows for the potential execution of compromised code in the development environment or canister.
- [CREDENTIALS_UNSAFE]: The skill facilitates the collection and storage of high-value secrets, including Twilio Account SIDs and API Key Secrets, in canister state. Handling these secrets while relying on a flagged external dependency significantly increases the risk of credential theft or exfiltration.
- [INDIRECT_PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection. Ingestion points: The sendSms function accepts to and body parameters (SKILL.md). Boundary markers: No delimiters or warnings are present for interpolated data. Capability inventory: The skill performs HTTP outcalls to the Twilio REST API via the twilio-client package. Sanitization: No escaping or validation of the message content is performed before transmission.
Recommendations
- AI detected serious security threats
- Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata