← 返回 Skills 市场
46
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install test-continuity
功能描述
Structured continuity and follow-up skill for OpenClaw agents. Use when an agent needs to decide whether natural user dialogue should stay as casual chat, be...
安全使用建议
Before installing or enabling this skill, do the following checks: 1) Verify the code—open and review scripts/personal_hooks.py and other scripts for network calls, file reads, or references to additional env vars; the main script is large (~1MB) and should be spot-checked for embedded data or unexpected behavior. 2) Confirm which fields under OPENCLAW_CONFIG_PATH the skill reads; avoid pointing it at a config that contains gateway tokens or unrelated credentials, or run the skill against a sanitized copy first. 3) Investigate the metadata mismatches (version and ownerId differences) with the publisher/source to ensure you have the intended release. 4) Run the regression harness and the install/setup steps in an isolated sandbox before connecting to production state. 5) If you plan to let the agent invoke this skill autonomously, restrict it initially (manual invocation only) until you audit runtime behavior and logs. If you need, ask the maintainer for a brief code walk-through or a smaller, auditable release that isolates the continuity-state logic from host config access.
功能分析
Type: OpenClaw Skill
Name: test-continuity
Version: 2.0.22
The skill bundle provides a sophisticated continuity and memory management framework for OpenClaw agents, allowing them to track user events, manage follow-up tasks, and maintain context across sessions. The included scripts (event_store.py, proactive_control.py, etc.) handle state persistence in the standard OpenClaw directory (~/.openclaw) and use keyword detection to categorize user intent. While the bundle includes instructions for the AI to perform setup tasks (such as following a wizard and deleting a setup file) and uses file manipulation utilities like send2trash, these behaviors are well-documented and strictly aligned with the stated purpose of providing a structured 'continuity engine.' No evidence of data exfiltration, malicious code execution, or harmful prompt injection was detected.
能力标签
能力评估
Purpose & Capability
The skill's name, README and SKILL.md describe a state-backed continuity/follow-up layer and the declared requirements (python3, OPENCLAW_STATE_DIR, OPENCLAW_CONFIG_PATH) are plausible for that purpose. However there are packaging inconsistencies (registry metadata shows version 2.0.22 while SKILL.md and _meta.json indicate 2.0.21, the ownerId values differ between registry metadata and _meta.json, and the registry label 'instruction-only' conflicts with the presence of numerous runnable scripts). These mismatches could be benign (release/version drift) but deserve verification.
Instruction Scope
SKILL.md's runtime instructions and boundaries stay within the stated continuity scope (init, setup, harness, guided settings). But the package includes large executable Python scripts (notably scripts/personal_hooks.py at ~1,079,911 bytes) which will be run locally if the operator follows README commands. The SKILL.md declares only OPENCLAW_STATE_DIR and OPENCLAW_CONFIG_PATH as required env vars; the codebase should be audited to confirm it does not read additional env vars, arbitrary host files, or contact unexpected external endpoints. The public docs warn not to publish live state, which implies the runtime will read/write sensitive host state — this increases the need for a code review before connecting to real state directories.
Install Mechanism
No automatic install spec is included in the registry entry (the package is 'instruction-only' in the registry), and installation is manual via copying the folder and running pip against requirements.txt or running provided scripts. There are no external download URLs or opaque installers in the registry metadata. This lowers automated supply-chain risk, but also means human operators must run install steps (and should inspect code first).
Credentials
The skill requests two environment/path inputs: OPENCLAW_STATE_DIR and OPENCLAW_CONFIG_PATH. These are reasonable for a stateful continuity layer, but OPENCLAW_CONFIG_PATH may point at a host config that contains unrelated secrets (gateway tokens, provider keys, chat-channel tokens). The SKILL.md does not explain exactly which config fields the skill reads. Because the skill operates on host state and may perform guided setup, requiring a path to a host config file is proportionally sensitive and should be validated: confirm the script reads only its own config subset and does not exfiltrate or indiscriminately read unrelated credentials.
Persistence & Privilege
The skill does not request always:true and is user-invocable; model invocation remains enabled (platform default). It does not declare system-wide modifications in SKILL.md. That privilege level is normal for a skill of this type. Still, because it is stateful and can be run by the agent, combine this with the other concerns (host config access, large runtime code) when deciding to allow autonomous runs.
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 test-continuity - 安装完成后，直接呼叫该 Skill 的名称或使用
/test-continuity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
v2.0.22
- Clarified and expanded documentation in SKILL.md, describing the skill's purpose, feature scope, runtime boundaries, and setup steps.
- Outlined V2 public scope, including context-sensitive memory classification, event chains, and deterministic first-run setup.
- Specified clear exclusions: no default companionship rhythm or generic proactive chatting.
- Provided detailed configuration and setup instructions.
- Listed entry points for developers, highlighting interaction with core scripts and functions.
元数据
常见问题
Test Continuity 是什么？
Structured continuity and follow-up skill for OpenClaw agents. Use when an agent needs to decide whether natural user dialogue should stay as casual chat, be... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 46 次。
如何安装 Test Continuity？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install test-continuity」即可一键安装，无需额外配置。
Test Continuity 是免费的吗？
是的，Test Continuity 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Test Continuity 支持哪些平台？
Test Continuity 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（darwin, linux）。
谁开发了 Test Continuity？
由 kylnwu（@kylnwu）开发并维护，当前版本 v2.0.22。
推荐 Skills