Security audit

酷家乐设计-免费版

Security checks for vulnerabilities and agentic risk

Overview

This skill appears to be an interior-design helper, but it asks users to handle an account token insecurely and run missing local scripts, so it needs review before use.

Review this before installing or using it. Do not put a real Kujiale access token in a project-root .kjlconfig.json or pass it on the command line; prefer an environment variable or secret manager. Because the referenced scripts are missing, do not download replacement scripts from an unauthenticated source or run files placed under ./scripts unless you have reviewed their origin and contents.

A.I.G

Vulnerability Patterns

Insecure DependenciesIntroduces malicious components through unsafe dependency sources
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)

T09 · Insecure Skill Coding Practices

Warning

Location: SKILL.md:145
Finding: Access Token Exposed Through Plaintext Configuration and Command-Line Arguments
Content: View full analysis
Remediation: View remediation

T08 · Insecure Dependencies

Note

Location: SKILL.md:222
Finding: Unpinned Third-Party Dependency Installation
Content: View full analysis
Remediation: View remediation

other

Note

Location: SKILL.md:84
Finding: Documented Workflow Invokes Scripts Missing From the Distributed Project
Content: View full analysis
--query="万科未来城" --areaId=<杭州id> --start=0 --num=20 node ./scripts/getFloorplanInfo.js --planId= node ./scripts/createDesign.js --token= --planId= node ./scripts/getTags.js --token= node ./scripts/getStyles.js --token= --tagItemIds= node ./scripts/triggerLayout.js --token= --designId= --tagIds= --styleId= --applyDecorationStyle=true --buildCeiling=true --autoDesign=true node ./scripts/trigger-render.js --obsDesignId= --xToken= ``` ### Technical Analysis The audited project contains only `SKILL.md`; the referenced `scripts/` directory and JavaScript files are absent. The package therefore cannot perform the documented API operations as distributed. Because the implementations are missing, the audit cannot verify API destinations, TLS certificate validation, upload handling, input validation, token redaction, authorization checks, or whether user-controlled parameters are safely passed to network requests. Relative script paths also create an integrity concern if users attempt to repair the package by downloading similarly named scripts from an unspecified source. The documentation does not provide an authenticated distribution location, checksums, signatures, or pinned revisions for the missing implementation. ### Attack Path 1. A user installs the Skill and attempts to follow the documented commands. 2. Execution fails because the referenced local scripts are not included. 3. The user searches for or receives replacement scripts from an external, unauthenticated source, or an untrusted party places files under the expected `./scripts` paths. 4. The user or Agent executes those scripts because their names mat ...[truncated 846 chars]
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands

Findings (7)

Vague Triggers

High

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The description says to use the skill for '需要设计创作、UI设计、海报制作、品牌视觉' even though the document otherwise describes an interior-design tool. These broad creative triggers overlap with common design requests and are not constrained by explicit activation boundaries or negative examples for those domains.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The manifest description says the skill should be used for 'UI设计、海报制作、品牌视觉', but the rest of the file consistently documents only Kujiale interior-design flows such as floorplan search, style selection, layout, and rendering. This is a semantic mismatch between claimed purpose and actual behavior, which appears limited to home/interior design operations.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The description explicitly says '支持中文交互', presenting a language constraint in the primary skill metadata. There is no accompanying opt-in, alternative language support, or clear justification that the skill must operate only in Chinese for compliance or regional reasons.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The trigger section states '需要设计创作、UI设计、海报制作、品牌视觉时使用', which is broad everyday language and does not define when the skill should or should not activate beyond a generic exclusion. This can cause accidental matching for many unrelated creative requests.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The skill explicitly tells users to create a local configuration file containing an access token and does not present a prominent warning at the point of setup. In an agent/tooling context with read/write/exec capabilities, plaintext credentials in project files are at elevated risk of accidental disclosure through source control, filesystem reads, debugging output, or later automation steps.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 82% confidence
Finding: The natural-language feature note highlights '中文交互' as a core capability, reinforcing a single-language expectation. Because no alternative language choice or opt-in is described nearby, this may conflict with language/locale neutrality expectations.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Low

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The setup flow instructs users to store an access token in a local JSON file, while the security section recommends environment variables instead. This contradiction increases the chance that users will persist secrets in plaintext files that may be committed, shared, or exposed through tooling, logs, or workspace access.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.