Back to skill

Security audit

README Sync

Security checks for vulnerabilities and agentic risk

Overview

This skill is a README maintenance helper whose file reads, project scanning, and README updates are disclosed and aligned with its purpose.

Before installing, confirm you are comfortable with the skill automatically reading README context during project work, creating a small pending-update JSON file, and proposing README.md changes. Review or disable auto-init behavior if you do not want a missing README.md generated from project structure.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (10)

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

The declared purpose frames the skill as a simple README sync helper, but the body expands behavior to directory scanning, codebase inference, README generation, and maintenance of an auxiliary state file. This mismatch is dangerous because users or calling agents may authorize the skill expecting limited documentation sync while it actually performs broader repository inspection and writes additional artifacts.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill documentation is written entirely in Chinese and does not indicate that the skill is intended only for Chinese-speaking users or provide any language/locale opt-in. Under the policy, forcing a specific language without user choice is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding

The skill describes file read/write behavior and even references helper scripts, but it declares no explicit tool scope such as permissions or allowed-tools. In an agent environment, that can cause the runtime or orchestrator to grant broader-than-intended filesystem access, making unintended reads or writes more likely if the skill is invoked automatically.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger conditions cover broad situations such as nearly any project-related question or code query/modification, which can cause the skill to run in many ordinary interactions without strong bounds. In practice, that increases the chance of unnecessary repository reads, auto-initialization attempts, or accumulation of pending write actions from loosely related conversations.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The workflow trigger language at this location uses ambiguous phrases like needing project background or related tasks, without precise limits on when the skill should activate. Ambiguous activation is risky in agent systems because it can expand the skill's reach from documentation support into default pre-task filesystem inspection.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The note states that project documentation should use Chinese by default, which imposes a language preference as a policy without requiring user choice. Although it mentions English projects may use English, it still defaults to a specific language rather than explicitly deferring to user or project preference.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

This code file contains natural-language instructions and output text primarily in Chinese, including the module docstring and many user-facing prompts, while the CLI description itself is in English. That creates a language/locale policy issue because the skill effectively enforces one language for usage and warnings without opt-in or justification.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The manifest describes a skill whose purpose is to read README before code work and update README afterward as needed, with confirmation required before writes. This file additionally implements full project scanning and automatic README initialization/generation (scan_project, auto_init, and init_readme), which goes beyond merely syncing README with ongoing work and materially expands the skill's behavior into repository summarization/bootstrap generation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

This markdown template forces a specific language for all headings and guidance text, which can violate a language/locale policy when no user opt-in or justification is provided. Because the file is a general README template rather than a clearly region-specific artifact, the fixed Chinese-only wording is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
75% confidence
Finding

When README.md is missing, the user-facing text suggests scan for scanning and auto-init for automatic scan-and-initialize. This partially blurs the distinction between read-only scanning and write behavior, and the surrounding documentation at the top presents these commands together under a README operations tool, even though auto_init later writes README.md. While not a strong contradiction, the inline guidance understates the side effect difference between scan and auto-init.

Content

No source excerpt is available for this finding.