← 返回 Skills 市场
context-recovery
作者
Jonathan Rhyne
· GitHub ↗
· v1.3.0
· MIT-0
6030
总下载
3
收藏
0
当前安装
7
版本数
在 OpenClaw 中安装
/install context-recovery
功能描述
Recover missing conversation context after explicit compaction or truncation, or when the user explicitly asks to recover prior work. Use for requests such as "where were we before compaction?" when the current thread is insufficient. Do not trigger on a generic "continue" when the current thread already provides an actionable next step.
安全使用建议
Before installing, understand that this skill helps an agent recover lost conversation context and may read recent thread history when needed. Its instructions are privacy-conscious and read-only by default, but users should still review any request to access another source before approving it.
能力评估
Credentials
The skill may access conversation history or approved external context sources, which can include private data, but it requires minimization, redaction, bounded reads, and user approval for separate sources.
Install Mechanism
The artifact is a single markdown skill file with no executable scripts, install hooks, or package behavior beyond skill instructions.
Instruction Scope
Activation is narrowly scoped to explicit recovery intent plus insufficient current context, with hard limits for current-thread history and explicit approval required before reading other sources.
Persistence & Privilege
The skill explicitly forbids persisting recovered summaries or extracted content unless the user consents to the exact redacted content and destination.
Purpose & Capability
The stated purpose is context recovery after compaction or truncation, and the instructions align with that purpose by limiting recovery to the smallest needed history and treating recovered material as untrusted evidence.
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 context-recovery - 安装完成后，直接呼叫该 Skill 的名称或使用
/context-recovery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
v1.3.0
Constrain activation and recovery scope, treat recovered material as untrusted, surface provenance and conflicts, and gate cross-source actions and persistence.
v1.2.2
Scanner hardening: add provenance fields, align _meta version, keep API-only recovery scope and consent-based persistence
v1.2.1
Hardening correction: publish API-first recovery instructions and openclaw metadata (fix previous publish path mismatch)
v1.2.0
Hardening pass: switch to openclaw metadata, remove broad filesystem/shell scan instructions, enforce API-first context recovery and consent-based persistence
v1.1.0
Fix suspicious flag: declare required bins (jq, rg, grep), add Safety Boundaries section documenting file access scope
v1.0.1
Fix: populate summary/description on ClawHub listing
v1.0.0
Initial release: Auto-recover context after session compaction. Adaptive depth fetching, thread-aware recovery, memory caching.
元数据
常见问题
context-recovery 是什么？
Recover missing conversation context after explicit compaction or truncation, or when the user explicitly asks to recover prior work. Use for requests such as "where were we before compaction?" when the current thread is insufficient. Do not trigger on a generic "continue" when the current thread already provides an actionable next step. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 6030 次。
如何安装 context-recovery？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install context-recovery」即可一键安装，无需额外配置。
context-recovery 是免费的吗？
是的，context-recovery 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
context-recovery 支持哪些平台？
context-recovery 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 context-recovery？
由 Jonathan Rhyne（@jdrhyne）开发并维护，当前版本 v1.3.0。
推荐 Skills