Security audit

multilingual-campaign

Security checks for vulnerabilities and agentic risk

Overview

The skill’s Canvora workflow fits its campaign-design purpose, with expected cloud processing and disclosed credit costs.

Use a trusted Canvora CLI and share only campaign material you are authorized to send to Canvora. Review the announced credit total: the workflow does not request separate price approval after confirming languages.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

The workflow sends campaign content to Canvora without an explicit third-party data disclosure.

Medium

Category: Output Handling
Confidence: 90% confidence
Finding: The prescribed generation command submits the campaign brief and brand identifier to Canvora, and the localization commands process the resulting visuals on that service. The skill explains costs and language confirmation but does not warn users that campaign content, potentially including unpublished or confidential material, is processed by a third party.
Content: No source excerpt is available for this finding.