任务收尾器
PassAudited by ClawScan on May 1, 2026.
Overview
This is a benign workflow-only skill, but it may make the agent keep progress logs and suggest autonomous follow-up steps.
This 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.
The agent may be more action-oriented and may propose or proceed toward follow-up work instead of stopping after a plan.
The skill intentionally changes the agent's response pattern by limiting planning-only replies, requiring execution evidence, and asking for an autonomous next step.
- 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
Use 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.
Task details could persist in logs or memory and later be reused as context.
The workflow references memory files and task history as places where goals, progress, and evidence may be recorded.
| 目标已陈述 | Select-String "目标" memory/{date}.md 匹配 |
...
- `memory/tasks.md` - 任务历史Keep evidence logs free of credentials, private data, and sensitive internal details; periodically review or clear persistent task records if needed.