Back to skill

Security audit

IRIS代码格式化专业版

Security checks for vulnerabilities and agentic risk

Overview

This is a Markdown-only IRIS/ObjectScript code review and formatting skill whose command examples are mostly purpose-aligned, but users should treat its broad operation wording as requiring explicit confirmation before changes.

Install only if you want an agent to help review and format IRIS/ObjectScript projects. Keep it pointed at explicit project directories, review generated shell commands before running them, and require a dry-run or confirmation before any formatting, reset, import, save, or delete-style action.

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (4)

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill is presented as a narrowly scoped code review/formatting tool, but its documented interface includes generic create/query/modify/delete and import/reset operations. That scope expansion increases the chance an agent will treat unrelated file or project state changes as in-scope, enabling unintended destructive actions under a trusted maintenance workflow.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill declares exec capability in a context that primarily needs file reading, pattern matching, and report generation. Exposing shell execution broadens the attack surface significantly because later instructions or user-controlled paths can be routed into command execution flows, especially in an agent environment that may trust tool-described workflows.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
86% confidence
Finding

The activation scope is defined by a long, fuzzy keyword list without meaningful boundaries, exclusions, or negative examples. In agent systems, this can cause over-triggering on unrelated prompts and inappropriate invocation of powerful tooling, increasing the risk of unintended file operations or misleading authority transfer to the skill.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill describes automatic create/modify/delete, import, and reset style operations without prominent warnings or consent requirements. In a code-governance tool, users may reasonably expect analysis-only behavior, so hidden or underexplained mutation capability can lead to accidental code changes, report overwrites, or destructive resets.

Content

No source excerpt is available for this finding.