Back to skill

Security audit

baeckerherz

Security checks for vulnerabilities and agentic risk

Overview

This skill provides fixed Baeckerherz brand information and includes a small local shell helper, with no evidence of hidden or harmful behavior.

This appears safe to install if you are comfortable granting a small brand-information skill command-execution permission. The current script only prints fixed Baeckerherz information, but a future update should be reviewed because the exec permission is broader than necessary for static documentation.

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
  • 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)

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill requests exec permission even though its documented behavior is limited to presenting static brand and feature information. Granting command-execution capability unnecessarily expands the attack surface: if the referenced script or surrounding environment is modified, the skill could run arbitrary shell commands under the agent's privileges.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest grants the skill "exec" permission and includes a shell script even though the declared purpose is only brand identity and feature overview information. This creates unnecessary command-execution capability, which expands the attack surface and could allow unintended local command execution if the skill or its entry content invokes the script.

Content

No source excerpt is available for this finding.