Security audit

Autobiographical Memory

Security checks for vulnerabilities and agentic risk

Overview

This memory skill is coherent, but it asks agents to persist and reuse personal context broadly without clear consent, scoping, or sensitivity controls.

Install only if you intentionally want an agent to keep cross-session personal memory. Before using it, define what may be stored, require confirmation for sensitive or inferred facts, keep secrets and regulated data out of memory, review MEMORY.md and daily notes regularly, and delete stale or unwanted entries.

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

Findings (1)

other

Warning

Location: SKILL.md:41
Finding: Excessive Personal Memory Access and Persistence Without Explicit Consent Boundaries
Content: View full analysis
Remediation: View remediation

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 (3)

Undeclared Tool Scope

Medium

Category: MCP Least Privilege
Confidence: 84% confidence
Finding: The skill clearly directs the agent to read local files such as MEMORY.md, USER.md, SOUL.md, TOOLS.md, and memory/YYYY-MM-DD.md, but it does not declare any explicit tool scope or permissions boundaries. That creates an authorization and review gap: an agent may access sensitive local memory files implicitly without a clear contract about what file access is required or allowed.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The skill description promotes persistent recording of conversations, decisions, preferences, and personal context across sessions, but it provides no privacy notice, consent requirement, minimization guidance, or retention warning. In a memory skill, that omission is especially dangerous because the normal operation of the skill is to accumulate potentially sensitive personal data over time, increasing exposure and compliance risk.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The instructions say to always write decisions, user preferences, important conversations, mistakes, and identity changes to daily notes, but they do not require filtering out sensitive personal data beyond a narrow example about API keys. This can lead to systematic storage of confidential or regulated information from conversations, making later unauthorized disclosure, over-collection, or secondary misuse more likely.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.