Back to skill

Security audit

小龙虾-DeepSeek版

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed DeepSeek coding-assistant skill that stores and resends conversation history, which is useful for its purpose but requires privacy caution.

Install only if you are comfortable with prompts, code, and prior conversation history being saved locally in plaintext and sent to DeepSeek on later requests. Clear history before discussing sensitive code, credentials, private logs, or regulated data.

Vulnerability Patterns
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
xiaolongxia.py:103
Finding

Plaintext Conversation Retention and Repeated External Disclosure

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • System Prompt LeakageDirect Leakage, Indirect Extraction, Tool-Based Exfiltration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (10)

Missing User Warnings

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill explicitly states that it will automatically carry forward conversation history and send it to the DeepSeek API, but it does not provide a clear warning or consent mechanism about privacy implications. This can expose sensitive source code, credentials, internal discussions, or personal data from prior turns to an external service without meaningful user awareness.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding

The skill advertises capabilities that imply file access, local persistence, and outbound network access, but it does not declare any explicit tool scope or permissions boundary. This weakens least-privilege controls and makes it harder for users or the platform to understand and constrain what the skill can read, write, or transmit.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger phrases are extremely broad and match common programming-related conversation, so the skill may activate in many contexts where the user did not intend to invoke an external coding assistant. Because the skill also sends prompts and history to a third-party API, accidental invocation can lead to unanticipated data exposure and overuse of privileged functionality.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill documents persistent storage of conversation history in a local file, but it does not warn users that sensitive prompts, code, secrets, or review content may be retained on disk. Local retention increases the chance of later disclosure through other users on the system, backups, malware, or unintended file access.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The system message is written entirely in Chinese and defines the assistant persona and output rules in Chinese, while the file contains no indication that users may choose another language. This creates a natural-language locale constraint without opt-in, which matches the language/locale policy violation category.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The description is written entirely in Chinese ("小龙虾-DeepSeek版超级编程助手") and does not indicate that users can choose another language or locale. Per the policy, forcing a specific language without opt-in is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The system prompt is entirely written in Chinese and defines the assistant persona and interaction rules in that locale, with no indication that the user can choose another language. This can violate language/locale policy when a skill imposes a specific language without opt-in or documented justification.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill persists full conversation history to a local file under the user's home directory without any notice, consent, retention policy, or access control hardening. Because this is a programming assistant, prompts may contain source code, secrets, tokens, internal URLs, or proprietary data, so silent storage materially increases privacy and secret-exposure risk on shared systems or if the account is later compromised.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill sends the entire message history, including prior user inputs, to the external DeepSeek API without an explicit warning or consent flow. In this skill's context, users are likely to paste code, logs, credentials, or confidential business data for debugging and review, so undisclosed third-party transmission can cause privacy, compliance, and secret-leak incidents.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

The file contains natural-language text only in Chinese ('DeepSeek 官方 API 配置'), which reflects a fixed language choice with no indication that users can select a preferred language. The policy explicitly calls for flagging language or locale constraints when they are imposed without opt-in or documented justification.

Content

No source excerpt is available for this finding.