← 返回 Skills 市场
372
总下载
1
收藏
0
当前安装
20
版本数
在 OpenClaw 中安装
/install openclaw-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
This package appears internally consistent with its stated purpose, but it contains sizeable runtime scripts bundled in the repo (e.g., scripts/personal_hooks.py ~1MB). Before installing: 1) Inspect the main scripts (personal_hooks.py, followup_skill_harness.py, install_local.sh) for any network calls, hardcoded endpoints, or code that reads environment variables other than the two declared ones. Search for uses of subprocess/os.system, socket/requests/urllib, smtplib/paramiko, eval/exec, or suspicious base64/obfuscation. 2) Do not point OPENCLAW_CONFIG_PATH at a production config that contains gateway tokens or provider keys; use a sanitized copy for initial tests. 3) Run the harness and init in a sandboxed OpenClaw workspace with an isolated state directory to verify behavior. 4) If you need higher assurance, run a quick static search for outbound networking and remote-execution patterns, and/or ask the maintainer for a short security summary of what the runtime reads/writes and any telemetry. If you find unexpected network access or secret-reading in the code, treat the package as suspicious and do not run it against production data.
功能分析
Type: OpenClaw Skill
Name: openclaw-continuity
Version: 2.0.21
The 'OpenClaw Continuity' skill bundle is a sophisticated framework designed to provide AI agents with time-aware continuity, memory staging, and follow-up capabilities. The core logic in scripts like `event_store.py` and `proactive_control.py` manages a structured system of 'events' and 'hooks' to track user topics across sessions, including support for '/new' session carryover. The `SKILL.md` provides comprehensive instructions for the agent to manage these states and perform guided configuration using natural language. While the bundle includes a file management utility (`file_output_sop.py`) that utilizes the `send2trash` library to move files to the system trash after processing, this behavior is documented as a manual verification step for assistant outputs. There is no evidence of malicious intent, data exfiltration, or unauthorized command execution; the bundle is well-documented, includes a robust test harness (`followup_skill_harness.py`), and its functionality aligns strictly with its stated purpose of enhancing agent memory and interaction continuity.
能力标签
能力评估
Purpose & Capability
Name/description (continuity / follow-up) align with declared runtime needs: python3 and two path env vars (OPENCLAW_STATE_DIR, OPENCLAW_CONFIG_PATH). No unrelated cloud credentials, tokens, or extra binaries are requested.
Instruction Scope
SKILL.md instructs running the included python scripts (init, setup-check, setup-apply, harness) and documents boundaries (frontstage vs internal state). It does not instruct the agent to read arbitrary host secrets or to send data to third-party endpoints in the provided text. Note: runtime scripts will read/write the declared state dir and the provided config path (which is expected for setup/validation).
Install Mechanism
No formal install spec in the registry (instruction-only), which is lowest install risk; README suggests pip install -r requirements.txt (single small dependency send2trash). There is an optional install_local.sh and packaging metadata (npm-style fetch). Because the package includes executable Python scripts (large personal_hooks.py), you should audit those files before executing them in sensitive environments.
Credentials
Only two required env vars (paths) are declared. That is proportionate for a state-backed skill. One caveat: OPENCLAW_CONFIG_PATH will point to a config file that may contain host tokens — avoid pointing it at live production config with credentials unless you have reviewed the code.
Persistence & Privilege
Skill isn't force-included (always: false) and uses normal autonomous invocation defaults. It does not declare system-wide config writes beyond its own state directory. The optional installer may create links in an OpenClaw workspace — review install_local.sh if you plan to run it.
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 openclaw-continuity - 安装完成后，直接呼叫该 Skill 的名称或使用
/openclaw-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
v2.0.21
OpenClaw Continuity 2.0.21
- Fixed ClawHub registry metadata parsing by using valid YAML frontmatter and the official metadata.openclaw.requires/envVars shape.
- Removed wording/SVG terms that could trigger unrelated crypto capability labels in the public bundle.
- No runtime behavior changes. Follow-up regression harness remains 14/14 pass; public upload remains portable and excludes host addons, web_live, local state, and private data.
v2.0.20
OpenClaw Continuity 2.0.20
- Improved documentation on required environment variables, including descriptions and sensitivity flags, for both generic and platform-specific metadata sections.
- No code changes; SKILL.md was updated to clarify environment setup and requirements.
- Enhanced clarity and consistency for setup validation across platforms (darwin, linux).
v2.0.19
2.0.19
- Move ClawHub runtime requirements to top-level frontmatter `requires` while keeping `metadata.openclaw` / `metadata.clawdbot` / `metadata.clawdis` aliases, so registry scanners can see python3 and OPENCLAW_STATE_DIR / OPENCLAW_CONFIG_PATH.
- No runtime behavior changes. Follow-up harness remains 14/14 pass.
- Public upload folder includes only portable skill files; host addons, web_live, local state, and private data are excluded.
v2.0.18
Broaden runtime requirements metadata for ClawHub scanners by declaring python3 and OPENCLAW_STATE_DIR / OPENCLAW_CONFIG_PATH at metadata top-level plus openclaw/clawdbot/clawdis aliases. No runtime behavior changes; harness remains 14/14 pass.
v2.0.17
Fix ClawHub runtime metadata format by using official metadata.openclaw YAML frontmatter for python3 and OPENCLAW_STATE_DIR / OPENCLAW_CONFIG_PATH requirements. No runtime behavior changes; harness remains 14/14 pass.
v2.0.16
Product documentation refresh with four setup/settings comic panels, complete feature coverage, guided setup details, natural-language and command setting examples, and portable skill-core packaging. Host/channel delivery remains adapter-owned; Telegram/TG is shown only as a minimal host-connected showcase.
v2.0.15
Align ClawHub top-level runtime metadata schema; keep portable-core ClawHub bundle and same-day wake handoff guard.
v2.0.14
Publish a portable-core ClawHub bundle while keeping host-only operator templates in GitHub source; includes the same-day wake handoff guard and aligned runtime metadata.
v2.0.13
Align ClawHub registry metadata with SKILL.md runtime requirements; includes the same-day wake handoff guard from v2.0.12.
v2.0.12
Fix routine-aware same-day wake handoff for night-owl schedules; add harness coverage, install placement diagram, and clearer natural-language settings guidance.
v2.0.10
Refined the public core package to avoid host-level config/session/cron path bindings while keeping contextual follow-up, routine-aware behavior, and in-chat controls intact.
v2.0.9
Copy-only refresh: clearer public positioning for in-chat controls, routine and do-not-disturb features, and contextual follow-up instead of generic cron-style messaging.
v2.0.8
Tightened the public core package to require an explicit OPENCLAW_STATE_DIR instead of autodiscovering workspace paths.
v2.0.7
Refined the English-first and Chinese-second public copy without changing the continuity core behavior.
v2.0.6
Clarify core runtime metadata for the public core-only package without changing continuity behavior.
v2.0.5
Repackaged the ClawHub distribution as a core-only continuity skill. Removed optional host-side addons and credential-oriented embedding paths from the ClawHub bundle while keeping the time-aware continuity core intact.
v2.0.4
**Major update: Host integrations, docs, and supplementary files removed. Core continuity logic now stands alone.**
- Removed 51 files including host-specific addons, documentation, and publishing assets.
- Now provides only the core time-aware continuity and follow-up logic for OpenClaw agents.
- Host-specific delivery, transport glue, and channel adapters are no longer bundled.
- Package description, name, and scope clarified as the standalone continuity core.
- Natural-language settings and core continuity features remain intact.
v2.0.3
v2.0.3
- corrected the public contact email to [email protected]
- fixed a frontstage leak where internal staging markers could appear in follow-up messages
- kept the shared core validation green (14/14 harness pass)
v2.0.1
English: corrected the public contact email to [email protected]; stabilized live checker delivery; prevented empty-provenance internal runtime replies from leaking to Telegram; improved release metadata and packaging.
中文：修正公開聯絡信箱為 [email protected]；穩定 live checker 發送路徑；避免無來源的 internal runtime reply 洩漏到 Telegram；同步更新發佈 metadata 與打包內容。
v2.0.0
OpenClaw Continuity 2.0.0 – Major Release
- Introduces structured continuity and follow-up for OpenClaw agents, with deterministic handling of chat, staged memory, and tracked follow-up.
- Adds support for managing parked topics, watchful states, delegated tasks, sensitive events, and candidate→incident→hook promotion.
- Implements explicit carryover, closure/cooldown, deduplication, dispatch cap, and daily-memory tracing for tracked items.
- Provides deterministic onboarding and guided-settings adjustment via natural language or explicit command triggers (e.g., `/care setup`).
- Clarifies runtime boundaries, with host-side responsibility for delivery, heartbeat glue, and transport integration.
- Expands language support for guided settings entry, covers explicit time zone handling, and ensures deterministic, state-backed continuity.
元数据
常见问题
OpenClaw 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 372 次。
如何安装 OpenClaw Continuity？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install openclaw-continuity」即可一键安装，无需额外配置。
OpenClaw Continuity 是免费的吗？
是的，OpenClaw Continuity 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
OpenClaw Continuity 支持哪些平台？
OpenClaw 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 OpenClaw Continuity？
由 Wakame（@redwakame）开发并维护，当前版本 v2.0.21。
推荐 Skills