Security audit

政企行政助手Gov Entity Admin

Security checks for vulnerabilities and agentic risk

Overview

This is a text-only Chinese administrative drafting skill with clear boundaries, though users should verify generated official records because examples sometimes add unsupported details.

Before installing, treat this as a drafting aid only. Require source notes for dates, attendees, owners, deadlines, decisions, metrics, and report claims, and review all generated administrative records before filing, sending, or relying on them.

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 (2)

T09 · Insecure Skill Coding Practices

Warning

Location: SKILL.md:102
Finding: Administrative meeting-minutes example fabricates unsupported official-record details
Content: View full analysis
2026-08-06 09:00, third-floor meeting room Director Wang (chair), Li, Engineer Zhang Access-control system acceptance The system is ready for acceptance, pending scheduling. Acceptance will be organized next Monday, 2026-08-10. Lead the access-control system acceptance Data-sharing interface funding Interface development requires dedicated funding, whose source is undetermined. Pause the work until the funding channel is confirmed. Determine the funding source for the data-sharing interface Year-end audit The audit scope and schedule require coordination. The office is instructed to lead preparation of the audit work plan. Prepare the year-end audit work plan ``` This example conflicts with the no ...[truncated 2249 chars]
Remediation: View remediation

T09 · Insecure Skill Coding Practices

Warning

Location: SKILL.md:131
Finding: Periodic-report example invents an entire administrative work record from an empty request
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (5)

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The documented activation examples are broad enough to match ordinary office-assistance requests, increasing the chance that the skill is invoked when the user did not explicitly intend to use it. In an administrative context, unintended activation can cause the model to apply the wrong workflow, output format, or boundary assumptions to unrelated tasks, which may lead to procedural mistakes or misleading formal documents.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The instruction explicitly says to use Chinese, which imposes a language choice on all interactions. Under the policy, forcing a specific language without user opt-in or a clearly documented justification is a natural-language policy violation.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 89% confidence
Finding: The manifest content is entirely written in Chinese, including the display name, description, and feedback text, with no indication that other languages are supported or that the user can opt into this locale. Under the policy, a skill that effectively enforces a specific language without documented choice can be a natural-language policy violation.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 85% confidence
Finding: All user-facing natural-language instructions in the README are Chinese, and there is no statement that the skill supports multiple languages or that Chinese is required due to region-specific constraints. Under the policy, forcing a specific language without user opt-in can be a natural-language policy issue.
Content: No source excerpt is available for this finding.

Vague Triggers

Low

Category: Not specified by scanner
Confidence: 84% confidence
Finding: The trigger says the skill activates when the user asks to draft, organize, or track the listed administrative documents and processes, but it does not provide explicit negative examples or tighter invocation constraints. In a manifest/markdown context, this broad phrasing may cause overlap with generic requests for writing or organizing workplace materials.
Content: No source excerpt is available for this finding.