Back to skill

Security audit

delegation-core

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill benign with medium confidence: its checks found a purpose-related capability, but flagged an MCP configuration step that is not disclosed in the description.

Before installing, consider whether you want a skill that may configure an MCP server in Claude in addition to delegating to external AI CLIs. The verdict is benign, not risk-free, and confidence is medium; Skillspector's analysis was partial because a referenced file was missing, although it inspected all six listed components.

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 (1)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The supplied command configures an MCP integration rather than delegating an execution task to an external AI CLI. Modifying Claude's MCP configuration and supplying authentication credentials are not disclosed by the description. The arguments are placeholders, so the snippet does not demonstrate successful configuration or exposure of real credentials.

Content

No source excerpt is available for this finding.