ClawHub
Back to skill

Security audit

JishuDB MCP Check / JishuDB MCP 验收助手

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed JishuDB connector test harness that uses scoped test fixtures and cleanup rather than hidden or unrelated behavior.

Use this only with a JishuDB connector you are authorized to test. The default smoke mode is read-only; all-tools mode will create temporary KB, document, note, and job data and then delete its own resources, so review the requested scope before approving it.

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
  • Rogue AgentSelf-Modification, Session Persistence
  • 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 (2)

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 215)May include surrounding context.

md
- [`assets/canary.pdf`](assets/canary.pdf): Use only in the all-tools fixture cases described by the test cases and fixture-values references; preserve exact by

Self-Modification

High
Category
Rogue Agent
Confidence
85% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · references/contract-baseline.md (reported line 68)May include surrounding context.

md
return no better than `PARTIAL`.
- A runtime-exposed digest differs: report the mismatch and return no better
  than `PARTIAL`, even when tool names are unchanged.
- Update this Skill version, test cases, expected counts, and digest whenever
  the accepted MCP contract changes.

Static analysis

No suspicious patterns detected.