Security audit

进度任务管理

Security checks for vulnerabilities and agentic risk

Overview

This is a task-tracking skill that persistently writes task progress files in the workspace, which matches its stated purpose but users should know it may record project details.

Install this if you want persistent workspace task tracking. Avoid putting secrets or sensitive business details in task descriptions or progress notes, and review TASKS.md, memory files, and .learnings/ before sharing the workspace.

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

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
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 (4)

Vague Triggers

Medium

Confidence: 95% confidence
Finding: The metadata sets the skill to always be available and the description advertises automatic updates on task state changes, which can cause the skill to activate and modify workspace files without an explicit, contemporaneous user request. In an agent environment, ambiguous or automatic activation increases the chance of unintended file writes, incorrect task creation, or persistence of user content into tracking files.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: These rules direct the skill to write to TASKS.md and memory/今天.md when a task is added or started, but the skill does not prominently warn users that ordinary task-related requests will mutate persistent workspace files. Hidden or insufficiently disclosed persistence is risky because users may share sensitive project details assuming they are ephemeral, while the skill stores them automatically.

Vague Triggers

Low

Confidence: 95% confidence
Finding: The phrase '给我一个明确的任务' is underspecified and can match ordinary conversational requests that the user did not intend to persist as formal tracked tasks. This ambiguity can lead to accidental task creation, incorrect status transitions, and unnecessary writes to persistent workspace files.

Missing User Warnings

Low

Confidence: 90% confidence
Finding: The memory integration section says important decisions, lessons, and cross-day progress will be recorded into persistent files, but it does not warn that potentially sensitive operational context may be retained beyond the current interaction. This can create unanticipated data retention and exposure risks, especially in shared repositories or long-lived workspaces.

Static analysis

No suspicious patterns detected.