Security audit

registry-sync

Security checks for vulnerabilities and agentic risk

Overview

The skill has a coherent registry-sync purpose, but it can automatically write metadata to a specific Feishu sheet or local backlog without explicit user confirmation.

Install this only if you control or intentionally use the named Feishu registry and the /Users/shiyi OpenClaw workspace paths. Before use, add a confirmation or dry-run step that shows the target tab, fields, and whether the write will be remote or queued locally.

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
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Vague Triggers

Medium

Confidence: 94% confidence
Finding: The description contains broad activation language ('use when... update, register, sync, or maintain') that can cause the skill to run in situations where the user did not explicitly intend external writes. Because this skill writes to an external Feishu spreadsheet and may also append to a local backlog file, unintended invocation can lead to unauthorized or incorrect data modification.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The skill instructs writes to a specific Feishu spreadsheet and fallback writes to a local backlog file, but the user-facing description does not clearly warn about these external and local side effects. Lack of transparent disclosure undermines informed consent and can cause users to expose project names, workflow details, or internal metadata to systems they did not expect to be modified.

Vague Triggers

Medium

Confidence: 97% confidence
Finding: The automatic trigger section instructs the agent to run this skill after creating new skills, templates, or workflows without waiting for user reminder. This creates an implicit side effect pathway where ordinary creation tasks can automatically propagate metadata to external systems or local files, increasing the risk of accidental disclosure, incorrect registration, or surprise writes.

Natural-Language Policy Violations

Low

Confidence: 81% confidence
Finding: The required metadata fields are presented only in Chinese, and the referenced spreadsheet/tab names are also language-specific, which implies a fixed locale for operation. The file does not state that this is a region- or organization-specific requirement or offer the user any language/locale choice.

Static analysis

No suspicious patterns detected.