Security audit

bidding-analyzer

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent bid-analysis assistant, but it asks for potentially confidential tender and company knowledge-base data without privacy or minimization guidance.

Review before installing if you handle confidential bids, pricing, customer details, certifications, or internal proposal material. Use sanitized excerpts or RAG snippets where possible, remove secrets and unnecessary personal or commercial data, and require human review before using generated bid content.

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

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
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
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (7)

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The README instructs users to upload tender documents and company knowledge bases, which commonly contain confidential commercial data, pricing, customer information, and internal qualification materials, but provides no warning about sensitivity, minimization, or redaction. In a procurement context, this omission increases the risk of unintended disclosure to the model platform, logs, downstream integrations, or unauthorized users handling the uploaded content.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: This YAML manifest describes a wide-ranging capability set but does not specify when the skill should be invoked, what user intents should trigger it, or any negative examples limiting scope. For manifest files, this creates an ambiguous activation surface that could overlap with many general document-analysis or proposal-writing requests.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The manifest description and workflow/output instructions are entirely Chinese and require a fixed Chinese markdown output schema, but there is no explicit user opt-in or documented region-specific justification. This can violate language/locale policy because the skill appears to enforce a single language by default rather than offering a choice.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The skill explicitly requests full bidding documents and company knowledge-base content, which commonly contain sensitive commercial, personal, and confidential procurement data, but provides no privacy handling, minimization, or warning guidance. In this context, users may paste proprietary bids, internal pricing, credentials, customer data, or regulated information into an LLM workflow without understanding retention, sharing, or redaction risks.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: This JSON file embeds natural-language skill examples, expected outputs, and validation requirements exclusively in Chinese. Because the file does not indicate that the skill is region-specific or provide any user opt-in or language choice, it appears to enforce a specific language/locale in a way that conflicts with the stated policy.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The natural-language description is entirely specified in Chinese and presents the skill behavior as Chinese-only, with no parameter or note allowing the user to choose another language or locale. This can violate language/locale policy requirements when users are not given an explicit opt-in or alternative.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 88% confidence
Finding: The natural-language instructions and examples in this file are written exclusively in Chinese, which can effectively force a specific language without user opt-in. Under the locale/language policy, this should either offer a language choice or clearly document and justify the language constraint.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.