Back to skill

Security audit

装修报价审核

Security checks for vulnerabilities and agentic risk

Overview

This is a Chinese renovation-quote review skill with no code or persistence, though users should redact quote documents before using any outside services it mentions.

Installers should understand this is advice-oriented renovation content, not executable automation. If using it with real quote photos or contracts, redact personal, address, payment, signature, QR-code, and contractor contact details before sharing with any AI service, app, social platform, or discussion group.

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
Findings (1)

other

Warning
Location
references/报价审核2025-2026新陷阱与审核新方法.md:26
Finding

Unwarned Disclosure of Renovation Quote Data to Third-Party Services

Content
View full analysis

Vulnerability Details

File Location: references/报价审核2025-2026新陷阱与审核新方法.md, line 26
Vulnerability Type: Privacy Data Exposure
Risk Level: Medium

Vulnerable Content

  1. Upload photographs of the quote to AI tools such as ChatGPT or ERNIE Bot for analysis.
  2. Use quote-audit features in renovation applications such as Haohaozhu or Yidoutang.
  3. Create a comparison table in Excel.
  4. Search social-media platforms for quotes from the same residential community and apartment layout.
  5. Ask users in renovation discussion groups.
    New method: AI assistance combined with manual review provides greater accuracy.

Technical Analysis

The reference material directs users to upload renovation quote photographs to external AI services and renovation applications and to discuss quote information through public or semi-public social channels. It does not instruct users to redact sensitive information, obtain consent from affected parties, assess service retention policies, or restrict the information disclosed.

Renovation quotes can contain personally identifiable and financially sensitive information, including customer names, telephone numbers, home addresses, apartment numbers, signatures, contractor identities, pricing details, account information, QR codes, and document identifiers. Uploading an unredacted image transfers this information outside the local Skill environment to services with separate access, retention, training, and redistribution policies.

The disclosure is avoidable because the Skill is designed to analyze documents supplied within its current environment. Recommending external uploads without a privacy warning creates an unnecessary data-exposure path.

Attack Path

  1. A user follows the reference guidance and photographs a renovation quote.
  2. The photograph includes personal, property, financial, or contractor information.
  3. The user uploads the unredacted photograph to an external AI service or re ...[truncated 1209 chars]
Remediation
View remediation

Remediation Suggestions

  1. Remove recommendations to share quote documents in public or semi-public discussion groups.
  2. Prefer analysis performed within the local Skill environment whenever possible.
  3. If use of an external service remains optional, add an explicit warning requiring users to redact:
    • Names and telephone numbers.
    • Full addresses, community names, building numbers, and apartment numbers.
    • Signatures and identity-document information.
    • Bank details, payment information, and QR codes.
    • Contract, customer, and document identifiers.
    • Contractor contact information and other third-party personal data.
  4. Require users to obtain consent before disclosing information belonging to contractors, household members, or other affected parties.
  5. Advise users to review the external service's data-retention, model-training, sharing, deletion, and privacy policies before uploading documents.
  6. Recommend sharing only the minimum fields required for analysis, such as item descriptions, quantities, unit prices, totals, material specifications, and construction methods.
  7. Where practical, provide a local redaction checklist or document-sanitization workflow before any external disclosure.
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (7)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The trigger list contains broad, common phrases such as '报价单', '装修预算', and '装修多少钱' that are likely to match routine user messages beyond a clear intent to invoke this skill. This can cause unintended activation, leading the agent to apply this skill in ambiguous contexts and potentially override more appropriate routing or expose reference-backed advice when the user did not explicitly request it.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

SQP-3 applies to all file types and covers natural-language policy violations such as forcing a specific language without user opt-in. This markdown file presents all guidance in Chinese and contains no indication that the skill is region-specific or that users can select another language.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

This markdown file is entirely written in Chinese and does not indicate that the user can choose another language or locale. Under the policy rule for natural-language violations, forcing a specific language without user opt-in is reportable when no justification or choice is provided.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The entire skill file is written only in Chinese and provides no indication that users may choose another language or that the content is intentionally limited to a Chinese-speaking audience. Under the policy, unqualified language forcing can be a locale-choice issue when no opt-in or justification is stated.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

This markdown file presents all instructions and guidance exclusively in Chinese, with no indication that users may choose another language or that the locale restriction is intentional. Under the policy rule, forcing a specific language without user opt-in 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
88% confidence
Finding

This markdown file contains user-facing instructional content exclusively in Chinese, and nowhere indicates that the skill is region-specific or that users may choose another language. Under the policy rule for natural-language constraints, forcing a single language without opt-in can be a locale-policy issue.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

This markdown file is entirely written in Chinese and provides no indication that language selection is optional or that the content is intentionally limited to a Chinese-speaking or region-specific audience. Under the policy rule, a skill artifact that effectively mandates a specific language without opt-in can be a natural-language policy concern.

Content

No source excerpt is available for this finding.