Ae1
High
- Category
- analysis-evasion
- Confidence
- 100% confidence
- Finding
Referenced artifact was not completely inspected
- Content
md node scripts/run.mjs --sample # 先看样例
Security audit
门店营收上报核对（免费版）
Security checks for vulnerabilities and agentic risk
Overview
This skill is a local Chinese-language store sales report checker that reads user-provided report text, runs deterministic calculations, and shows no evidence of network use, persistence, credential access, or destructive behavior.
Install only if you are comfortable running a local Node script on store sales report text. The skill appears local and non-persistent, but it is Chinese-oriented and expects reports with specific Chinese headers; review the listed free-version omitted checks before relying on its output.
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 (8)
Ae1
High
- Category
- analysis-evasion
- Confidence
- 100% confidence
- Finding
Referenced artifact was not completely inspected
- Content
md node scripts/run.mjs --sample # 先看样例
Description-Behavior Mismatch
Medium
- Category
- Not specified by scanner
- Confidence
- 76% confidence
- Finding
The manifest description says the free version executes only the engine-declared free check items, implying a constrained subset. But later documentation states the bundled engine runs locally and that the only difference from the full edition is the specific omitted checks, which effectively claims the rest of the full engine behavior is present. That is a semantic mismatch in how the scope of actual behavior is described.
- Content
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 93% confidence
- Finding
This file embeds user-facing descriptions, advice, and contract text solely in Chinese, which effectively forces a specific language for users interacting with the skill. The policy allows locale constraints only when users are given a choice or when the restriction is clearly documented and justified; neither is present here.
- Content
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 96% confidence
- Finding
This file includes natural-language strings that force a specific language/locale for all users, including the header comment, usage text, error messages, and runtime output. Under the policy, a fixed language is a violation unless the skill offers user choice or clearly documents a justified region-specific constraint, which this file does not.
- Content
Intent-Code Divergence
Low
- Category
- Not specified by scanner
- Confidence
- 92% confidence
- Finding
The skill promises local-only, no-network behavior, but declares Bash as an allowed tool. Bash can invoke arbitrary shell utilities, including network-capable commands such as curl, wget, nc, ssh, or language runtimes, so the manifest grants capabilities broader than the user-facing trust claims. In a skill handling business reporting data, that mismatch could enable unintended data exfiltration or execution paths outside the documented safety model.
- Content
Intent-Code Divergence
Low
- Category
- Not specified by scanner
- Confidence
- 94% confidence
- Finding
The inline documentation says non-JSON input is treated as plain-text material such as a pasted '合同全文' (full contract text), which conflicts with the surrounding manifest and usage text that consistently limit the skill to store sales report tables with headers. This is an intent/documentation mismatch rather than an implementation necessity, because the actual tool messaging and capability are specific to 门店营收上报表核对.
- Content
Natural-Language Policy Violations
Low
- Category
- Not specified by scanner
- Confidence
- 88% confidence
- Finding
The JSON contains only Chinese-language headers and labels in the sample text, indicating a fixed locale-specific format. For a general-purpose skill artifact, this can violate language/locale policy when no user opt-in or justification for Chinese-only output is provided.
- Content