Security audit

Tunnel Vision & Slack

Security checks for vulnerabilities and agentic risk

Overview

This time-management skill needs review because its instructions leave recurring calendar changes and automatic meeting declines insufficiently controlled.

The advisory calendar concern is supported by the workflow's direct-action wording, although no calendar integration or executable automation is bundled. Static scanning reported no findings, and the artifacts show no malicious intent. Use the skill for planning first; before applying changes, agree on the calendar, slots, end date, invitation exceptions, and how to disable automatic declines.

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

Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
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)

The workflow directs recurring calendar blocking and automatic meeting declines without warning about their effects on existing scheduling and invitations.

Medium

Category: Output Handling
Confidence: 90% confidence
Finding: These are steps in the skill's own workflow, not merely a catalog of available tools. Creating recurring blocks and enabling auto-decline can alter calendar availability and reject invitations, but the skill does not warn the user about those consequences or require approval before applying changes.
Content: No source excerpt is available for this finding.