v1.0.2

任务收尾器

BenignClawScan verdict for this skill. Analyzed May 1, 2026, 6:12 AM.

Analysis

This is a benign workflow-only skill, but it may make the agent keep progress logs and suggest autonomous follow-up steps.

GuidanceThis skill is mainly a task-discipline template. It is reasonable to install if you want clearer goals, progress updates, and completion evidence. Before using it, decide whether the agent may take autonomous follow-up actions, and instruct it not to record credentials or sensitive information in evidence, memory, or task-history files.

Findings (2)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

Abnormal behavior control

Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.

Agent Goal Hijack

SeverityLowConfidenceHighStatusNote

SKILL.md

- 仅计划的回复：最多 1 次
- 下一次回复必须包含执行证据
...
**NEXT_AUTONOMOUS_STEP**:
- 一个无需用户输入即可执行的后续步骤

The skill intentionally changes the agent's response pattern by limiting planning-only replies, requiring execution evidence, and asking for an autonomous next step.

User impactThe agent may be more action-oriented and may propose or proceed toward follow-up work instead of stopping after a plan.

RecommendationUse this skill when you want stricter task completion discipline, and explicitly require confirmation before any follow-up step that changes files, accounts, deployments, or public content.

Sensitive data protection

Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.

Memory and Context Poisoning

SeverityLowConfidenceHighStatusNote

SKILL.md

| 目标已陈述 | Select-String "目标" memory/{date}.md 匹配 |
...
- `memory/tasks.md` - 任务历史

The workflow references memory files and task history as places where goals, progress, and evidence may be recorded.

User impactTask details could persist in logs or memory and later be reused as context.

RecommendationKeep evidence logs free of credentials, private data, and sensitive internal details; periodically review or clear persistent task records if needed.