Back to skill

Security audit

Instinct平替:个人琐事委托与办结回执

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed Chinese-language personal-errand workflow skill with an optional AI-HIVE connection and a read-only result checker; I found no hidden execution, persistence, or exfiltration.

Install this only if you are comfortable using a Chinese-language workflow that may process errands, orders, billing, subscriptions, and travel details. If you connect AI-HIVE, confirm the domain, permissions, upload scope, pricing, currency, and budget first; do not paste passwords, tokens, verification codes, cookies, or full identity documents, and require separate approval for payments, cancellations, messages, refunds, deletions, or external writes.

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
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (9)

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The declared description presents an end-user personal assistant workflow skill for delegating errands and completing practical tasks with authorized tools. The supplied code instead implements an offline result-record checker. Its primary purpose is validation of JSON output against a contract, including status/action invariants, evidence requirements, authorization references for mutating actions, and local artifact hash/path checks. It explicitly avoids live actions, API calls, and authentication. This is a materially different purpose and capability set from the declared assistant functionality, so the description does not accurately represent the code.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The workflow materially diverges from the declared skill purpose: it performs subscription billing analysis, renewal tracking, and cancellation guidance rather than the manifest’s stated errands, call-prep/follow-up, and shopping-comparison functions. This scope mismatch is dangerous because it can cause the agent to request or process unexpected financial/account data, defeating user and platform expectations and weakening permission and risk review boundaries.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
96% confidence
Finding

The workflow content is materially out of scope for the declared skill purpose: it handles travel itinerary disruption and booking-change follow-up rather than the manifest’s stated personal errand/receipt assistant scenarios. This kind of capability mismatch is dangerous because routing or safety controls may rely on the manifest, causing users or orchestrators to invoke an unexpected workflow that processes different sensitive data, performs different actions, or bypasses expected review boundaries.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

This markdown file presents all user-facing instructions in Chinese and does not indicate that language selection is optional or configurable. That can violate language/locale policy when users are not explicitly given a choice.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

This markdown file contains user-facing operational instructions exclusively in Chinese, and there is no indication that the user can opt into another language or that the skill is intentionally limited to a Chinese-speaking context. Under the language/locale policy rule, forcing a specific language without user choice can be a natural-language policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The entire skill is written as Chinese-only operational guidance and does not indicate that language selection is optional or that the workflow is intentionally restricted to Chinese-speaking users or a China-specific compliance context. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The file's natural-language instructions and examples are entirely in Chinese, and there is no indication that users can opt into another language or that the workflow is intentionally restricted to a Chinese-language audience. Under the policy rule for language/locale constraints, forcing a specific language without user choice is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The JSON examples are entirely written in Chinese in both the input and expected behavior fields, which indicates the skill behavior is specified for a single language. There is no visible indication that users may choose another language or that the locale restriction is intentional and justified for a region-specific use case.

Content

No source excerpt is available for this finding.