Security audit

Quality Verification

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed quality-checking helper that runs local diagnostics, optional tests, and simple session metrics for development workflows.

Install this for trusted development workspaces where you are comfortable with local lint/test commands running from hooks. Review repository test scripts before enabling TEST_BEFORE_COMMIT, and consider adding explicit capability metadata plus session ID validation for stricter deployments.

SkillSpector

Vulnerability Patterns

MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
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)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 92% confidence
Finding: The skill declares no explicit permissions, yet its documented behavior clearly relies on shell execution, environment variables, and reading/writing files via hooks and state files. This creates a transparency and governance gap: operators may enable the skill without understanding that it can run commands like linters/tests, inspect repository files, and write artifacts such as bracket.json, which increases the risk of unintended command execution or data exposure.

Static analysis

No suspicious patterns detected.