Back to skill

Security audit

git-history-leak-check-lite

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious because it reports inspecting Git metadata beyond its stated patch-only secret scan; the scan is partial, so the evidence is incomplete.

Consider the reported extra Git metadata access before installing. If you need only a scan of supplied patch lines, wait for complete review or choose a tool whose behavior stays within that scope; these results do not establish that the artifact is safe.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (2)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

The supplied-diff-only scope does not disclose the surrounding repository metadata access described in the finding, creating a limited privacy and least-privilege violation. Although upgrade recommendations are explicitly documented, that does not disclose or authorize additional filesystem inspection to generate them; there is no evidence of network transmission or credential exfiltration.

Content

No source excerpt is available for this finding.

The patch scanner inspects surrounding Git metadata solely to generate an upgrade recommendation.

Medium
Category
Excessive Agency
Confidence
96% confidence
Finding

The scanner searches ancestor directories, follows a .git file's gitdir target, and reads logs/HEAD despite needing only the supplied patch to perform its advertised task. This unnecessarily accesses potentially unrelated metadata and exposes a derived activity count in the report; an attacker-controlled metadata target could also cause an oversized or blocking synchronous read, although no network exfiltration occurs.

Content

No source excerpt is available for this finding.