Ae1
High
- Category
- analysis-evasion
- Confidence
- 100% confidence
- Finding
Referenced artifact was not completely inspected
- Content
md node scripts/run.mjs --sample
Security audit
投标报价AI审查（免费版）
Security checks for vulnerabilities and agentic risk
Overview
This is a local bid-quote arithmetic checker with a minor scope-description mismatch, but no hidden network, persistence, credential, or destructive behavior was found.
Before installing, note that the free version only checks line-item arithmetic and missing quantity, unit-price, or amount fields. Do not rely on the English description's broader feature list unless a later version aligns the docs and code. Use it only on bid quote files you intend to process locally, and treat the output as an aid rather than an official bid evaluation.
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 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
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (9)
Ae1
High
- Category
- analysis-evasion
- Confidence
- 100% confidence
- Finding
Referenced artifact was not completely inspected
- Content
md node scripts/run.mjs --sample
Ae1
High
- Category
- analysis-evasion
- Confidence
- 100% confidence
- Finding
Referenced artifact was not completely inspected
- Content
md node scripts/run.mjs --sample
Ae1
High
- Category
- analysis-evasion
- Confidence
- 100% confidence
- Finding
Referenced artifact was not completely inspected
- Content
md 脚本本身不做计算，真正的检查在同目录的 `scripts/engine/quote-audit.js`（纯 Node 标准库，CommonJS），可以直接引用：
Ae1
High
- Category
- analysis-evasion
- Confidence
- 100% confidence
- Finding
Referenced artifact was not completely inspected
- Content
md 脚本本身不做计算，真正的检查在同目录的 `scripts/engine/quote-audit.js`（纯 Node 标准库，CommonJS），可以直接引用：
Description-Behavior Mismatch
Medium
- Category
- Not specified by scanner
- Confidence
- 96% confidence
- Finding
L013 says the free skill audits totals, the tender upper price cap, and written-figure cross-checks. Elsewhere in the manifest, the free version is explicitly limited to line-item arithmetic and omission prompts, and it explicitly says those additional checks are not included. This creates a semantic mismatch in the skill description about what the free version actually does.
- Content
Intent-Code Divergence
Low
- Category
- Not specified by scanner
- Confidence
- 89% confidence
- Finding
L064 states '脚本本身不做计算' ('the script itself does not do calculations') while immediately identifying
scripts/engine/quote-audit.jsas the component that performs the real checks. Because that engine is part of the skill code being referenced for direct use, the wording is misleading and contradicts the surrounding implementation description rather than merely omitting detail.- Content
Natural-Language Policy Violations
Low
- Category
- Not specified by scanner
- Confidence
- 90% confidence
- Finding
The code hard-codes
toLocaleString('en-US', ...), which imposes a specific locale format for output regardless of the user's language or regional preferences. This matches the policy category for language/locale constraints because no opt-in or justification is provided in the file.- Content
Natural-Language Policy Violations
Low
- Category
- Not specified by scanner
- Confidence
- 94% confidence
- Finding
The JSON value contains only Chinese text and provides no indication that the user can choose a language or locale. Under the stated policy, forcing a specific language without opt-in is a natural-language policy concern.
- Content