← 返回 Skills 市场
aghareza

taskwarrior

作者 Ali Aghareza · GitHub ↗ · v0.1.3
cross-platform ✓ 安全检测通过
793
总下载
1
收藏
2
当前安装
4
版本数
在 OpenClaw 中安装
/install taskwarrior
功能描述
Workspace-local task management powered by Taskwarrior. Add, organize, and track tasks by project, tags, due dates, and priority with all data stored inside...
安全使用建议
This skill is a thin wrapper around the Taskwarrior CLI and appears to be what it claims: it requires the 'task' binary and will store all data under <workspace>/.openclaw/taskwarrior/. Before installing/use: 1) Ensure you trust the environment image (that 'task' is the official Taskwarrior binary). 2) Confirm you are comfortable with the skill writing task state into the workspace directory (it will not touch your global ~/.task or ~/.taskrc unless you explicitly ask). 3) If you do not want the agent to perform writes autonomously, consider restricting autonomous invocation or making the skill user-invocable only. 4) Note the small metadata mismatch around install options (apt vs brew) — confirm your runtime provides Taskwarrior rather than relying on the skill to install it.
功能分析
Type: OpenClaw Skill Name: taskwarrior Version: 0.1.3 The OpenClaw AgentSkill for Taskwarrior is designed with strong security principles. It explicitly confines all Taskwarrior data and configuration to a workspace-local directory (`<workspace>/.openclaw/taskwarrior/`), preventing any interaction with global user configurations. The `SKILL.md` and `references/safe_command_policy.md` files contain clear instructions for the AI agent to avoid dangerous operations like `delete` or `purge` without explicit user request, and to preview bulk changes. The skill does not perform system-level installations at runtime and only uses standard Taskwarrior commands. There is no evidence of data exfiltration, unauthorized execution, persistence mechanisms, or prompt injection attempts designed to subvert the agent for malicious purposes.
能力评估
Purpose & Capability
Name and description match the requested artifact: the skill requires the 'task' (Taskwarrior) binary and performs only Taskwarrior operations scoped to the active workspace. Requested binaries and declared capabilities are proportional to the stated task-management purpose.
Instruction Scope
SKILL.md instructs only Taskwarrior commands and workspace-local file operations (creating .openclaw/taskwarrior/, writing a taskrc there, using TASKRC/TASKDATA env vars). It does not read unrelated system files or exfiltrate data to external endpoints. It explicitly forbids writing global ~/.task unless explicitly requested.
Install Mechanism
No bundled code is present; install metadata references package managers (brew/apt) which is reasonable for obtaining the Taskwarrior binary. The skill itself states it will not perform runtime installs. This is coherent for a CLI-wrapper skill, though there is a small metadata mismatch (SKILL.md metadata includes apt and brew, while registry install spec lists only brew) — a minor documentation inconsistency, not a security issue.
Credentials
The skill requires no credentials or sensitive environment variables. It optionally reads standard workspace-related env vars (OPENCLAW_WORKSPACE, WORKSPACE, PROJECT_DIR, REPO_ROOT) to resolve the workspace root — reasonable and scoped. No unexpected secrets or unrelated services are requested.
Persistence & Privilege
always:false and no modifications to other skills or system-wide settings. The skill writes only to the workspace-scoped .openclaw/taskwarrior/ area (its own data), which is appropriate for a task-management skill. Autonomous invocation is the platform default but is not in itself a red flag here.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install taskwarrior
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /taskwarrior 触发
  4. 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
v0.1.3
- Added skill metadata specifying required binaries and supported installation methods (apt, brew). - No functional changes to core behavior or workflow.
v0.1.2
- Added a YAML frontmatter section with skill name and description to improve metadata clarity. - No other functional or behavioral changes.
v0.1.1
- Switched to require preinstalled Taskwarrior in the environment; the skill no longer attempts to install Taskwarrior itself. - Clarified dependency checks and error messaging for missing Taskwarrior. - Strengthened guarantees that no global Taskwarrior config or data are touched unless explicitly requested. - Expanded documentation for workspace-local storage layout, runtime requirements, usage, and safety guidelines. - Improved and clarified operation workflows and supported commands.
v0.1.0
Initial release of Taskwarrior skill. - Manages tasks using Taskwarrior with data stored per workspace. - Automatically detects Taskwarrior and attempts installation if needed. - Supports common operations: add, list, modify, mark done, projects, tags, due dates, priorities, annotations. - Stores task data at <workspace>/.openclaw/taskwarrior/ using workspace root detection. - Prevents task deletion/purge unless specifically requested. - Ensures no global ~/.task modifications.
元数据
Slug taskwarrior
版本 0.1.3
许可证
累计安装 2
当前安装数 2
历史版本数 4
常见问题

taskwarrior 是什么?

Workspace-local task management powered by Taskwarrior. Add, organize, and track tasks by project, tags, due dates, and priority with all data stored inside... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 793 次。

如何安装 taskwarrior?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install taskwarrior」即可一键安装,无需额外配置。

taskwarrior 是免费的吗?

是的,taskwarrior 完全免费(开源免费),可自由下载、安装和使用。

taskwarrior 支持哪些平台?

taskwarrior 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 taskwarrior?

由 Ali Aghareza(@aghareza)开发并维护,当前版本 v0.1.3。

💬 留言讨论