Back to skill

Security audit

Soft Landing — Reflect Without Pressure

Security checks for vulnerabilities and agentic risk

Overview

This is a low-risk reflection skill that only guides the agent to create gentle day or week review summaries from user-provided text.

Before installing, be aware that the skill may steer vague reflection-like requests into a structured day, yesterday, or week review. It does not appear to access private files, run code, use credentials, or persist data.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill advertises common phrases like “look back on yesterday” and “help me make sense of today,” which are broad enough to overlap with normal conversation outside an explicit request to invoke a reflection workflow. This can cause unintended activation, leading the agent to steer the user into a structured reflection flow when the user may have meant a casual conversational request.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill explicitly treats vague phrases such as “I want to reflect a bit,” “can we do a review,” and “help me process things” as candidates for this skill, only asking a timeframe clarification. Because these phrases are common and underspecified, the skill may hijack generic emotional support or planning conversations and route them into this specialized behavior without sufficient intent confirmation.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.