other
Warning
- Location
setup.md:32- Finding
Sensitive Built-In Agent Memory Can Be Duplicated into a Secondary Plaintext Store
- Content
View full analysis
- Remediation
View remediation
Security audit
AI Agent Memory
Security checks for vulnerabilities and agentic risk
Overview
This skill is a local memory organizer, but it can persist and duplicate sensitive personal or work context without enough safeguards or deletion guidance.
Review before installing if you may store private contacts, business decisions, client information, or anything sensitive. Use it only for information you are comfortable keeping as local plaintext files, avoid secrets and regulated data, and confirm exactly what will be copied before enabling built-in memory sync.
Vulnerability Patterns
- Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
- 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
Findings (2)
T09 · Insecure Skill Coding Practices
Warning
- Location
SKILL.md:14- Finding
Potentially Sensitive Memory Data Is Stored in Plaintext Without Explicit Permission Hardening
- Content
View full analysis
- Remediation
View remediation
Vulnerability Patterns
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (8)
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 94% confidence
- Finding
The skill encourages persistent storage of projects, contacts, decisions, and knowledge on the local filesystem but does not present a prominent warning that data will be written to and modified under ~/memory. Users may disclose sensitive or regulated information without realizing it will be durably stored and later searchable, increasing privacy, compliance, and data exposure risks if the machine or account is compromised.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The invocation examples are extremely broad (for example, generic phrases like 'Save this...' or 'Find...') and provide no confirmation or scoping boundaries before performing file writes, searches, or updates in persistent storage. In an agent setting, this can cause unintended triggering from normal conversation context, leading to unauthorized persistence, modification of notes, or retrieval of sensitive locally stored information.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 92% confidence
- Finding
The template explicitly encourages storing persistent contact and relationship data such as names, roles, companies, communication preferences, and interaction history, but provides no guidance on handling sensitive or personal information. In an agent memory skill, this increases the likelihood that users or downstream agents will retain unnecessary personal data indefinitely, creating privacy, compliance, and unintended disclosure risks if the memory store is synced, shared, or accessed by other tools.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 92% confidence
- Finding
The documented sync workflow instructs copying data from built-in memory into user-managed files without any warning about sensitive, personal, or system-derived information. That creates a real privacy and data-governance risk because built-in memory may contain confidential preferences, personal data, or other context that should not be broadly persisted, duplicated, or exposed to weaker access controls.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The skill proposes syncing built-in memory into a separate external memory system without a clear privacy warning or explicit consent flow tailored to copying previously shared personal data. That is dangerous because it expands the scope and persistence of retained data, potentially duplicating sensitive preferences, contacts, or decisions into local files the user did not realize would be created.
- Content
Ssd 3
Medium
- Category
- Not specified by scanner
- Confidence
- 94% confidence
- Finding
The setup defines a straightforward data-retention workflow: copy prior memory, solicit new information, and persist it long-term in an organized external system. In context, this is not merely descriptive; it operationalizes ongoing retention of potentially sensitive personal and professional data without minimization, retention limits, or sensitivity screening, increasing privacy and confidentiality risk.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The skill instructs the agent to create and populate files under ~/memory/ and immediately store user information there, but it does not require explicit informed consent, disclosure that local filesystem writes will occur, or any caution about storing sensitive personal data. In a memory-management skill, this creates a real privacy and data-handling risk because users may disclose personal or confidential information without understanding it will be persisted in local files.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 89% confidence
- Finding
The troubleshooting guide instructs users to manually copy content from the agent's built-in memory into ~/memory/sync/ but does not warn that built-in memory may contain sensitive, private, or access-controlled information. This can lead to unintended duplication of confidential data into a broader or less protected storage area, increasing the risk of leakage, over-retention, or accidental exposure.
- Content