← 返回 Skills 市场
Periodic Reflection
作者
jpengcheng523-netizen
· GitHub ↗
· v1.0.0
· MIT-0
110
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install periodic-reflection
功能描述
周期性反思报告生成工具。用于自动化生成结构化的自我进化反思报告,支持多场景(EvoMap 发布、Agent 进化、DevOps 运维等)。 **触发场景:** - 用户要求生成周期性反思报告 - 需要量化指标对比和版本追踪 - 需要数据驱动的优化决策 - 用户提到"反思"、"复盘"、"进化报告"、"周期性总结"...
安全使用建议
What to consider before installing or scheduling this skill:
- The included script (scripts/generate-report.js) simply writes a markdown report with hardcoded example metrics and does not make network calls or read arbitrary system files — running it in a test directory is low risk. Try: node scripts/generate-report.js --output /tmp/test-reflection.md to confirm behavior.
- The documentation instructs you to create or edit a metrics-collector that runs shell commands against logs and other scripts. Those collectors are not bundled. If you add or use a metrics collector, review any commands it will run carefully — shell commands or external scripts can access any local files and could be used to exfiltrate data if malicious.
- Missing files: metrics-collector.js and version-manager.js are referenced but not provided. That is an inconsistency; ask the author or maintainer for the missing components and review them before use.
- Cron risk: do not add to crontab for unattended execution until you (a) review/replace any metric-collecting code, (b) run the skill in a sandboxed environment, and (c) ensure outputs and logs go to an expected directory.
- Additional checks: inspect any scripts you add for network calls, external endpoints, or use of sensitive paths; run the tool in a container or non-privileged account first; confirm the package source/maintainer (no homepage given) and prefer installing only from a known repository or after manual review.
功能分析
Type: OpenClaw Skill
Name: periodic-reflection
Version: 1.0.0
The periodic-reflection skill bundle is a legitimate utility designed to generate automated performance reports and 'evolution' summaries for AI agents or DevOps tasks. The core logic in scripts/generate-report.js is limited to local file system operations (reading/writing Markdown reports) and lacks any network connectivity, sensitive data access, or obfuscated code. While the documentation in SKILL.md suggests using crontab for automation and mentions shell commands for metric collection, these are standard administrative practices for the tool's stated purpose and do not exhibit malicious intent or hidden backdoors.
能力评估
Purpose & Capability
The declared purpose (periodic reflection / report generation) matches the included main script (scripts/generate-report.js) which generates markdown reports. However, SKILL.md and the architecture diagram reference additional components (scripts/metrics-collector.js, scripts/version-manager.js, calc-* scripts, logs, domain-history.json) that are not present in the package. That mismatch suggests the docs describe a larger system than what's actually shipped.
Instruction Scope
Runtime instructions tell users to edit a metrics-collector to run shell commands (e.g. 'cat logs/publish.log | wc -l', 'node scripts/calc-*.js') and to schedule the generator with cron. The shipped generate-report.js does not execute arbitrary commands or access remote endpoints (it uses a local collectMetrics() that returns hardcoded values), but the documentation encourages adding command-based collectors that, if later executed, could run arbitrary shell/node commands against local files. This is scope-creep risk if users import or implement metric collectors without review.
Install Mechanism
No install spec is provided (instruction-only with a small JS script). Nothing is downloaded or installed automatically by the skill, minimizing install-time risk.
Credentials
The skill declares no required environment variables, no credentials, and no config paths. All file I/O in the provided script is limited to writing a report under a reports/ path. The documentation references logs and config files, but none of those require secrets or external credentials to use the included code.
Persistence & Privilege
The skill does not request permanent presence (always: false) and does not modify other skills or system settings. The recommended cron usage is normal for a report generator but implies care before scheduling.
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install periodic-reflection - 安装完成后,直接呼叫该 Skill 的名称或使用
/periodic-reflection触发 - 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
Initial release of periodic-reflection.
- Automates generation of structured periodic reflection reports for EvoMap publishing, Agent evolution, and DevOps scenarios.
- Supports data-driven indicator tracking, version comparison, and changelog solidification.
- Includes scripts for report generation, metrics collection, and version management.
- Provides configurable monitoring metrics and sample report templates.
- Features recommended reflection cycles and best practices for continuous improvement.
- Guides on indicator definitions, action planning, and exception handling mechanisms.
元数据
常见问题
Periodic Reflection 是什么?
周期性反思报告生成工具。用于自动化生成结构化的自我进化反思报告,支持多场景(EvoMap 发布、Agent 进化、DevOps 运维等)。 **触发场景:** - 用户要求生成周期性反思报告 - 需要量化指标对比和版本追踪 - 需要数据驱动的优化决策 - 用户提到"反思"、"复盘"、"进化报告"、"周期性总结"... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 110 次。
如何安装 Periodic Reflection?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install periodic-reflection」即可一键安装,无需额外配置。
Periodic Reflection 是免费的吗?
是的,Periodic Reflection 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
Periodic Reflection 支持哪些平台?
Periodic Reflection 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 Periodic Reflection?
由 jpengcheng523-netizen(@jpengcheng523-netizen)开发并维护,当前版本 v1.0.0。
推荐 Skills