Back to skill

Security audit

Lobster Autodream

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed memory-consolidation helper that persists selected conversation notes, with privacy and frequency limits described, but users should understand it can update long-term memory automatically.

Install this only if you want the agent to summarize conversations into long-term memory. Review or disable the HEARTBEAT.md integration if you do not want automatic background updates, and periodically inspect MEMORY.md for sensitive or stale entries.

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)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill is designed to run during heartbeat checks and session compression, which are broad background events rather than narrowly scoped user actions. Because the invocation boundary is not tightly constrained, the agent may perform memory extraction and persistent writes at times the user did not clearly intend, increasing the risk of accidental retention of sensitive or irrelevant conversation data.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The user-trigger includes a very short alias, "autodream," alongside a generic phrase for memory organization, making accidental or ambiguous activation more likely in normal conversation. In a skill that reads conversation history and updates long-term memory, unintended activation can cause privacy-impacting persistence of data without a deliberate user request.

Content

No source excerpt is available for this finding.

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
80% confidence
Finding

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Content

Scanner excerpt · SKILL.md (reported line 113)May include surrounding context.

markdown
## AutoDream
- 检查是否需要整理记忆
- 参照 skills/autodream/SKILL.md

注意事项