Security audit

agentgates-inference

Security checks for vulnerabilities and agentic risk

Overview

The skill offers paid external AI inference with clearly disclosed payments and user-controlled spending.

Send only content you intend Agentgates to process, and check its privacy and attestation assurances before submitting sensitive material. Approve payments individually or set a spending budget; protect wallet credentials and bearer keys.

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 inference workflow sends prompt content to an external service without a privacy warning.

Medium

Category: Output Handling
Confidence: 91% confidence
Finding: The workflow instructs the agent to submit the request body, including chat messages, to api.agentgates.ai. Although the skill distinguishes confidential and open model tiers, it does not warn that submitted messages leave the user's environment or caution against including sensitive information, particularly when using the unattested open tier.
Content: No source excerpt is available for this finding.