Back to skill

Security audit

Memory Compactor

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed memory organizer, but it can automatically rewrite or remove persistent memory without clear confirmation, preview, or rollback controls.

Review this before installing if you rely on persistent memory. Enable it only if you are comfortable with automatic memory summarization and cleanup, and prefer a version that previews changes, asks for confirmation, and keeps recoverable backups.

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 (3)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill explicitly describes compressing and cleaning memory files, including removing expired or one-time information, but does not warn the user that stored memory content may be modified or deleted. In a memory-management skill, unattended alteration of persistent data can cause irreversible loss of important context, preferences, or records if the compaction logic is overbroad or inaccurate.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill schedules automatic execution at fixed times and on threshold triggers, meaning it may alter stored memory without contemporaneous user awareness or approval. This increases the risk of silent data loss, unwanted rewriting of long-term memory, and difficulty auditing what changed after the fact.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
80% confidence
Finding

Most user-facing descriptions and trigger phrases are written in Chinese, including the stated manual trigger phrase, with no indication that other languages are supported or that Chinese is an optional locale. This can constitute a language/locale policy issue when a specific language is imposed without user opt-in.

Content

No source excerpt is available for this finding.