Back to skill

Security audit

Chuyao Aa

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent MTL API wrapper generator, but its examples encourage credential handling and proxy endpoints that would need careful review before use.

Only install or use this skill if you will adapt the generated code to your application's security model: load MTL credentials from a secrets manager or protected runtime config, require authentication and authorization on every generated route, restrict exposed MTL operations to an allowlist, avoid returning raw exception messages, and redact or omit request and response payloads from logs.

Vulnerability Patterns
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (3)

T05 · Unauthorized Access and Privilege Escalation

Error
Location
SKILL.md:49
Finding

Generated MTL Proxy Endpoints Lack Authentication and Authorization

Content
View full analysis
getAlterSheet(@RequestParam("id") String id) { Map result = MapBuilder.of("success", true).build(); RequestMessage request = RequestMessage.builder() .method(HttpMethod.GET) .path("/dev/api/v1/alterSheet/detail") .json(null) .parameters(MapBuilder.ofAssignType("id", id).build()) .build(); try { ResponseMessage responseMessage = mtlClient.sendRequest(request); if (null == responseMessage) { throw new RuntimeException("response is null"); } result.put("data", LubanTypeUtil.toJSON(responseMessage.getJson())); } catch (Exception e) { result.put("success", false); result.put("msg", e.getMessage()); } return result; } ``` ### Technical Analysis The skill directs the agent to generate Spring MVC endpoints that forward requests through a server-held `ApiClient`. No authentication annotation, authorization policy, caller identity validation, role check, or resource-level access check is included. Consequently, any caller capable of reaching the generated `/mtl` route may be able to exercise the MTL permissions assigned to the application's client identity. This creates a confused-deputy condition in which an unprivileged caller causes the application to perform privileged operations on the ca ...[truncated 1581 chars]
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
SKILL.md:29
Finding

MTL Credentials Are Intended to Be Embedded in Java Source Code

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:146
Finding

Optional Logging Captures Complete MTL Requests and Responses Without Redaction

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep