Security audit

Algernon Review

Security checks for vulnerabilities and agentic risk

Overview

This flashcard review skill is purpose-aligned, but its raw SQLite instructions can let unescaped user or AI-generated text alter the local study database in unintended ways.

Review this skill before installing if your OpenAlgernon study database contains important or sensitive material. It should be revised to use parameterized SQLite operations, validate slugs and IDs, escape generated text safely, and document or make optional the conversation-log append.

A.I.G

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 (1)

T09 · Insecure Skill Coding Practices

Error

Location: SKILL.md:35
Finding: Unsafe SQL Interpolation in Database Operations
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The skill persists review-session details to disk without any notice or consent mechanism, which is a privacy and transparency failure. Because review sessions may reflect user interests, study content, timestamps, and performance metrics, silent storage can surprise users and create compliance or confidentiality issues, especially in shared or synced environments.
Content: No source excerpt is available for this finding.

Context-Inappropriate Capability

Low

Category: Not specified by scanner
Confidence: 90% confidence
Finding: The skill writes session metadata into a persistent conversation log even though its stated purpose is flashcard review and scheduling. This creates unnecessary secondary storage of user activity data, expanding the privacy surface and potentially exposing study habits or sensitive topics if the memory directory is later accessed or reused by other components.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.