← 返回 Skills 市场
1468
总下载
2
收藏
19
当前安装
36
版本数
在 OpenClaw 中安装
/install autonomous-tasks
功能描述
Self-driven AI worker. Reads goals, generates tasks, executes, and logs progress. Keywords: create goal, new goal, set goal, run goals, 创建目标, 新目标, 设定目标, 执行目标.
使用说明 (SKILL.md)
Autonomous Tasks
Read goals → Generate tasks → Execute → Log → Stop
You are a self-driven AI worker. Each time you are woken up, execute one round of tasks, then stop.
All user data lives in agents/ relative to this SKILL.md file's directory (i.e. the same directory that contains this SKILL.md). This data is preserved across normal skill updates (only SKILL.md and _meta.json are overwritten).
Workflow
0. Initialize (first-time only)
If agents/ does not exist (relative to this SKILL.md's directory):
- Ask the user for their goals
- Read
assets/templates.mdand create all files inagents/ - Strongly recommend the user to set up scheduled execution:
openclaw cron add --name "autonomous-tasks" --message "run autonomous tasks" --every 1h - Stop immediately. Do not continue to the next steps. Wait for the next wake-up.
1. Read Goals
Read the following files from agents/ (relative to this SKILL.md's directory):
agents/AUTONOMOUS.md— long-term goals + current todosagents/memory/backlog.md— backlog ideasagents/memory/tasks.md— unfinished tasks from a previous run
If current todos are empty, check milestones:
- If there are unchecked milestones
[ ]: take the next one, decompose it into concrete todos, write them into the "Current Todos" section of AUTONOMOUS.md, then continue - If all milestones are done: prompt the user to set new goals. Give 2-3 example directions based on project context. Once the user has set new goals, clean up old state:
- Clear completed milestones from
AUTONOMOUS.md - Clear
memory/backlog.md - Clear
memory/tasks-log.md - Do not invent goals. If the user doesn't respond, stop and wait
- Clear completed milestones from
2. Generate Tasks
If memory/tasks.md has unfinished tasks, resume execution without regenerating.
If no unfinished tasks, generate new tasks from todos and write to memory/tasks.md:
- [ ] task description
- [ ] task description
Rules:
- Prioritize
AUTONOMOUS.mdcurrent todos first, thenbacklog.md - Split into reasonable granularity, each task must have a clear output
- All outputs go to the current working directory, never into the skill directory or
agents/ - Keep outputs from different goals and milestones separated
3. Execute Tasks
Execute tasks in order from memory/tasks.md.
Mark as in progress:
- [~] task description
Mark as done:
- [x] task description → output path
If execution fails, mark and skip:
- [!] task description → failure reason
Do not retry failed tasks.
If you discover new ideas or follow-up work during execution that is not part of the current task, add it to memory/backlog.md instead of acting on it immediately.
4. Archive
When all tasks in memory/tasks.md are marked ([x] or [!]):
- Append results to
memory/tasks-log.md:
- ✅ description → output path (YYYY-MM-DD)
- ❌ description → failure reason (YYYY-MM-DD)
- Clear
memory/tasks.md(keep the heading) - Remove completed items from
AUTONOMOUS.mdorbacklog.md - If all current todos are cleared, mark the corresponding milestone as
[x] - When
tasks-log.mdexceeds 50 lines, keep only the most recent 30
5. Stop
After archiving, stop immediately. Do not generate new tasks. Do not loop. Wait for the next wake-up.
Reference
Before starting, read assets/rules.md (same directory as this SKILL.md) for prohibited actions, core principles, and file structure.
安全使用建议
This skill appears to do what it says (run self-driven tasks), but its runtime instructions are too vague about what 'execute tasks' means. Before installing or scheduling it: (1) Review or create agents/AUTONOMOUS.md, agents/memory/tasks.md and backlog to ensure no sensitive goals or commands are present. (2) Run the skill in a restricted/sandboxed environment (isolated working directory, no access to secrets, limited network) to observe what it does. (3) Do not enable the suggested cron schedule until you trust the behavior. (4) Ask the author or update SKILL.md to explicitly constrain what kinds of task execution are allowed (e.g., only edit files, no shell commands, whitelist network domains, require explicit user confirmation for external actions). (5) Avoid placing the skill's working directory somewhere that contains secrets or production data. These steps will reduce risk from its broad, under-specified execution powers.
功能分析
Type: OpenClaw Skill
Name: autonomous-tasks
Version: 10.3.1
The skill bundle implements a structured autonomous task management system that reads goals from local markdown files and executes them in a controlled loop. It includes explicit safety rules in 'assets/rules.md' to prevent self-modification or unauthorized git operations, and the suggested persistence via 'openclaw cron' is transparently presented to the user as a functional recommendation.
能力评估
Purpose & Capability
Name/description match the SKILL.md: it reads goals, generates tasks, executes them, and logs progress in an agents/ directory. No unrelated environment variables or external dependencies are requested. However, 'execute tasks' is under-specified — legitimate implementions might only operate on text files, but the instructions permit arbitrary task execution and writing outputs to the current working directory, which is broader than a simple task-planning helper.
Instruction Scope
The SKILL.md explicitly instructs the agent to read and write files under agents/ (relative to the skill) and to 'execute tasks' from memory/tasks.md, but it does not define what executing a task entails or limit allowed actions (shell commands, network calls, API keys, etc.). It also recommends installing a scheduled cron job (openclaw cron add). The guidance to never write into the skill dir but allow arbitrary outputs to the current working directory grants broad write/read scope that could touch sensitive locations if the working directory is not sandboxed.
Install Mechanism
Instruction-only skill with no install spec and no code files — minimal install risk. Nothing is downloaded or written by an installer step.
Credentials
The skill requests no environment variables or credentials, which is proportionate. It does, however, expect persistent filesystem access to create and maintain an agents/ directory adjacent to SKILL.md and to write outputs to the working directory; those filesystem privileges may be more than a user expects from a 'task generator' and should be considered when granting runtime permissions.
Persistence & Privilege
always:false (good) and the skill is user-invocable. It recommends creating a scheduled cron entry via openclaw, which would make it run periodically if the user sets that up. Because disable-model-invocation is false (the platform default), the skill could act autonomously when invoked — combine this with vague execution semantics and persistent storage and the blast radius increases.
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 autonomous-tasks - 安装完成后，直接呼叫该 Skill 的名称或使用
/autonomous-tasks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
v10.3.1
Extract first-time setup into step 0 with explicit stop, cleaner workflow flow
v10.2.5
Republish
v10.2.4
- Added new documentation files under agents/ for agent memory and goals tracking.
- Updated setup instructions for clarity and consistency.
- Slightly reworded recommendations related to scheduling and first-time setup flow.
- No changes to workflow logic or core functionality.
v10.2.3
Strongly recommend scheduling after first-time setup
v10.2.2
Extract templates and rules to assets/ for lazy loading, add stop after first-time setup
v10.2.1
Store user data in agents/ relative to SKILL.md directory instead of <workspace> placeholder
v10.3.0
Replace <workspace> placeholder with ./ (cwd-relative paths) so AI can resolve file paths without clawhub variable injection
v10.2.0
**Big change: Workspace file location and separation clarified**
- All user data now resides in `<workspace>/autonomous-tasks/` instead of the workspace root.
- Skill never writes to its own directory; all outputs must go to `<workspace>/`, outside the `autonomous-tasks/` folder.
- Updated all instructions, file paths, and examples for the new directory structure.
- No workflow or feature changes beyond path and file handling adjustments.
v10.1.2
- Updated description to include “create goal, new goal, set goal, run goals” and their Chinese equivalents for improved discoverability
- No workflow or functionality changes; all operational details remain the same
- Version bump to 10.1.2 reflecting metadata update only
v10.1.1
- Added more descriptive keywords to the skill’s description for improved discoverability.
- No workflow or logic changes; documentation remains consistent with the previous version.
- Version bump to 10.1.1.
v10.1.0
- Bump version to 10.1.0 in metadata for consistency.
- No functional or workflow changes to the skill logic or documentation.
v10.0.0
**Major update: All user data and logs now live in the workspace directory, not the skill directory.**
- All files tracking goals, tasks, backlog, and logs have been moved to the user’s workspace, keeping the skill directory clean and update-safe.
- Skill workflow is refactored to read/write only user data in the workspace; the skill directory now contains only SKILL.md and _meta.json.
- Skill no longer initializes or mutates user files inside its own directory.
- Updated onboarding: asks for workspace path and goals on first run; auto-creates workspace files from templates.
- Removed memory files and AUTONOMOUS.md from the skill directory to avoid accidental data loss during upgrades.
v6.2.1
- Add guidance: If you discover new ideas or follow-up work during task execution, add them to backlog.md instead of acting immediately.
- Clarifies that only items related to the current task should be executed; unrelated ideas go to the backlog.
v6.2.0
- Task numbering was removed: tasks now use plain descriptions instead of "TASK-XXX" IDs.
- All references to numbered task IDs and their incrementing were deleted from workflow steps and examples.
- Task result lines in `tasks-log.md` no longer show numbered IDs; they simply use the task description.
- The rest of the workflow and file structure remain unchanged.
v6.1.4
- Refined user prompting when all milestones are complete: prompts for new goals and waits for user input before cleaning up old state.
- Clarified that cleanup of milestones, backlog, and task log only occurs after new goals are set.
- Emphasized not inventing goals and to stop if user does not respond.
- No functional changes to task execution or file handling.
v6.1.3
- Update milestone completion routine: when all milestones are done, now clear completed milestones, backlog, and task log, and reset task IDs before prompting for new goals.
- Revised instructions for end-of-milestone workflow and workspace reset.
- No changes to code or external interface; SKILL.md documentation only.
v6.1.2
- Added explicit instruction to check if the workspace path is set in AUTONOMOUS.md before generating tasks.
- Now prompts the user to configure the workspace if not set, and stops execution until configured.
- If the workspace directory does not exist, it will be created before task generation.
- Updated file structure section to clarify separation between the skill directory and the workspace.
v6.1.1
- Updated first-time setup instructions: after completing all milestones, now prompts the user to set new goals and a new workspace path.
- Clarified: on finishing all milestones, user is prompted for a new workspace path as well as new goals.
- No behavioral changes to task or output handling.
v6.1.0
- All outputs must now be written to a user-defined workspace directory specified in AUTONOMOUS.md, not into the skill directory.
- First-time setup now guides the user to configure a workspace path.
- Prohibited actions updated: writing outputs in the skill directory is explicitly disallowed; outputs must go to the workspace.
- Clarified that outputs from different goals/milestones should be kept separated within the workspace.
v6.0.0
**English workflow and documentation rewrite**
- Rewrote all documentation in SKILL.md from Chinese to English for broader accessibility.
- No workflow or behavior changes—logic and best practices remain as before.
- All file structure, guide steps, and prohibited actions are now described in English.
- Skill purpose and usage instructions are unchanged.
元数据
常见问题
autonomous-tasks 是什么？
Self-driven AI worker. Reads goals, generates tasks, executes, and logs progress. Keywords: create goal, new goal, set goal, run goals, 创建目标, 新目标, 设定目标, 执行目标. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 1468 次。
如何安装 autonomous-tasks？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install autonomous-tasks」即可一键安装，无需额外配置。
autonomous-tasks 是免费的吗？
是的，autonomous-tasks 完全免费（开源免费），可自由下载、安装和使用。
autonomous-tasks 支持哪些平台？
autonomous-tasks 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 autonomous-tasks？
由 glimmer（@changye01）开发并维护，当前版本 v10.3.1。
推荐 Skills