Security audit

Raspberry Pi Manager

Security checks for vulnerabilities and agentic risk

Overview

This is advertised as a Raspberry Pi manager, but the inspected files show a local plaintext logging/export tool that does not actually manage devices.

Review this before installing if you expected real Raspberry Pi administration. Treat it only as a local operations journal, do not enter credentials or sensitive host details, and remember that entries are stored in plaintext under ~/.local/share/raspberry-pi-manager and may be searched or exported.

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

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (9)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 99% confidence
Finding: The skill’s manifest and title present it as a Raspberry Pi device-management tool, but the documented behavior is actually a generic local logging/export utility. This kind of description-behavior mismatch is dangerous because users may supply operationally sensitive data under the assumption that the tool performs device actions locally and safely, when it instead persists plaintext notes and exports unrelated to the claimed functionality.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The manifest advertises Raspberry Pi management, while the body describes a note-taking/audit system that logs arbitrary user input to local storage. This deceptive packaging can lead users or downstream agents to invoke the skill in inappropriate contexts and disclose credentials, hostnames, incident details, or infrastructure notes that are then retained and exportable.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

High

Category: Not specified by scanner
Confidence: 99% confidence
Finding: The script’s implemented behavior materially contradicts the advertised Raspberry Pi management purpose and instead acts as a generic local data-collection and logging utility. In a skill ecosystem, this kind of capability mismatch is dangerous because users or agents may grant trust and provide operational/device-related input under false pretenses, enabling covert collection and later export of sensitive data.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The command documentation implies actions like connect, monitor, automate, webhook, and status relate to real Raspberry Pi operations, but the documented implementation only records user-provided text. In context, this can mislead operators into believing a device was monitored or managed when nothing occurred, creating security and reliability risks through false audit trails and missed operational actions.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The skill documents persistent logging, global history, search, and export of operational data, but does not warn that inputs may contain sensitive infrastructure details and will be stored in plaintext under the user’s home directory. In a Raspberry Pi or fleet-operations context, this increases the risk of exposing credentials, IPs, maintenance details, or incident records to other local users, backups, or accidental exports.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The explicit 'social tool/social toolkit' wording inside a skill labeled as a Raspberry Pi manager is a strong indicator of deceptive packaging or repurposed code. This discrepancy increases the likelihood that the skill is misrepresented to obtain trust, causing users to disclose information or invoke actions they would not otherwise permit.
Content: No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The presence of unrelated commands such as webhook, template, schedule, notify, connect, and analytics expands capability beyond the claimed Raspberry Pi management context without justification. In this context, such extraneous features are suspicious because they create additional avenues for collecting, organizing, and redistributing user-provided data under a misleading device-management label.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: The export function silently aggregates all previously stored log contents into a new file, increasing the persistence and portability of potentially sensitive user data. This amplifies privacy and data-handling risk because information that was already retained locally becomes easier to copy, share, or exfiltrate without a dedicated confirmation or warning.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: These command handlers persist arbitrary user-provided input into per-command log files and a history log without any clear notice, retention policy, or consent flow. This is dangerous because users may enter hostnames, credentials, internal URLs, operational notes, or other sensitive content expecting ephemeral processing, but the script stores it locally for later retrieval and aggregation.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.