Back to skill

Security audit

Sqlite Lite Manager Free

Security checks for vulnerabilities and agentic risk

Overview

This SQLite management skill is local and purpose-aligned, but it asks for write and exec authority while giving broad activation guidance and unsafe database mutation examples without clear safeguards.

Install only if you want an agent to help manage local SQLite files. Before allowing it to run write, delete, cleanup, migration, or exec-driven database commands, require the agent to identify the exact database file, preview affected rows, make or confirm a backup, and get explicit confirmation for destructive actions.

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
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (6)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The activation guidance is broad enough to trigger on generic requests involving databases, SQL, or storage management, which can cause unintended invocation of a skill with read/write/exec capabilities. In an agent environment, overbroad routing increases the chance that the skill is used in contexts the user did not intend, potentially leading to unauthorized file or database modifications.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill explicitly supports create/modify/delete database operations but does not pair those capabilities with user-facing warnings, confirmation requirements, or backup guidance. In the context of an agent that may act autonomously with write/exec access, this raises the risk of accidental data loss or destructive changes to local databases.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The overview states the tool provides '自动备份' and 'Schema迁移等常用能力', which reads as available capabilities of this skill. Later, the free-version limitations explicitly say scheduled automatic backup and schema auto-migration are not supported, creating a direct documentation contradiction about what the current skill edition actually does.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The keyword coverage list includes many fragmented and overly generic trigger terms, making accidental activation more likely. Because the skill advertises write and exec tools, unintended activation can expose local data stores to destructive queries, schema changes, or command execution in response to loosely related prompts.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The examples demonstrate deletion and cleanup operations as normal usage without emphasizing irreversibility, scoping, or confirmation safeguards. In an AI-agent context, examples strongly shape behavior, so unsafe examples can normalize destructive execution against the wrong table, row set, or database file.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The dependency section says the skill is purely local SQLite, requires no API key, and does not involve cloud service calls. The error-handling table later includes '网络错误' and instructs users to test connectivity with ping and check firewalls/proxies, which contradicts the stated local-only operating model.

Content

No source excerpt is available for this finding.