# Send Autonomous Skill to your agent
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
## Fast path
- Download the package from Yavira.
- Extract it into a folder your agent can access.
- Paste one of the prompts below and point your agent at the extracted folder.
## Suggested prompts
### New install

```text
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete.
```
### Upgrade existing

```text
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run.
```
## Machine-readable fields
```json
{
  "schemaVersion": "1.0",
  "item": {
    "slug": "autonomous-skill",
    "name": "Autonomous Skill",
    "source": "tencent",
    "type": "skill",
    "category": "AI 智能",
    "sourceUrl": "https://clawhub.ai/feiskyer/autonomous-skill",
    "canonicalUrl": "https://clawhub.ai/feiskyer/autonomous-skill",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadUrl": "/downloads/autonomous-skill",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=autonomous-skill",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "packageFormat": "ZIP package",
    "primaryDoc": "SKILL.md",
    "includedAssets": [
      "SKILL.md",
      "scripts/run-session.sh",
      "templates/executor-prompt.md",
      "templates/initializer-prompt.md",
      "templates/task-list-template.md"
    ],
    "downloadMode": "redirect",
    "sourceHealth": {
      "source": "tencent",
      "slug": "autonomous-skill",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-04-30T00:34:30.890Z",
      "expiresAt": "2026-05-07T00:34:30.890Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=autonomous-skill",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=autonomous-skill",
        "contentDisposition": "attachment; filename=\"autonomous-skill-1.0.3.zip\"",
        "redirectLocation": null,
        "bodySnippet": null,
        "slug": "autonomous-skill"
      },
      "scope": "item",
      "summary": "Item download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this item.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/autonomous-skill"
    },
    "validation": {
      "installChecklist": [
        "Use the Yavira download entry.",
        "Review SKILL.md after the package is downloaded.",
        "Confirm the extracted package contains the expected setup assets."
      ],
      "postInstallChecks": [
        "Confirm the extracted package includes the expected docs or setup files.",
        "Validate the skill or prompts are available in your target agent workspace.",
        "Capture any manual follow-up steps the agent could not complete."
      ]
    }
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/autonomous-skill",
    "downloadUrl": "https://openagent3.xyz/downloads/autonomous-skill",
    "agentUrl": "https://openagent3.xyz/skills/autonomous-skill/agent",
    "manifestUrl": "https://openagent3.xyz/skills/autonomous-skill/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/autonomous-skill/agent.md"
  }
}
```
## Documentation

### Autonomous Skill - Long-Running Task Execution

Execute complex, long-running tasks across multiple sessions using a dual-agent pattern (Initializer + Executor) with automatic session continuation.

### Directory Structure

All task data is stored in .autonomous/<task-name>/ under the project root:

project-root/
└── .autonomous/
    ├── build-rest-api/
    │   ├── task_list.md
    │   └── progress.md
    ├── refactor-auth/
    │   ├── task_list.md
    │   └── progress.md
    └── ...

This allows multiple autonomous tasks to run in parallel without conflicts.

### Workflow Overview

User Request → Generate Task Name → Create .autonomous/<task-name>/ → Execute Sessions

### Step 1: Initialize Task Directory

Generate a task name from user's description and create the directory:

# Generate task name (lowercase, hyphens, max 30 chars)
# Example: "Build a REST API for todo app" → "build-rest-api-todo"
TASK_NAME=$(echo "$USER_TASK" | tr '[:upper:]' '[:lower:]' | sed 's/[^a-z0-9]/-/g' | sed 's/--*/-/g' | cut -c1-30 | sed 's/-$//')

# Create task directory
TASK_DIR=".autonomous/$TASK_NAME"
mkdir -p "$TASK_DIR"

echo "Task directory: $TASK_DIR"

### Step 2: Analyze Current State

Check if this is a new task or continuation:

TASK_DIR=".autonomous/$TASK_NAME"

# Look for existing task list
if [ -f "$TASK_DIR/task_list.md" ]; then
  echo "=== CONTINUATION MODE ==="
  echo "Found existing task at: $TASK_DIR"

  # Show progress summary
  TOTAL=$(grep -c '^\\- \\[' "$TASK_DIR/task_list.md" 2>/dev/null || echo "0")
  DONE=$(grep -c '^\\- \\[x\\]' "$TASK_DIR/task_list.md" 2>/dev/null || echo "0")
  echo "Progress: $DONE/$TOTAL tasks completed"

  # Show recent progress notes
  echo ""
  echo "=== Recent Progress ==="
  head -50 "$TASK_DIR/task_list.md"
else
  echo "=== NEW TASK MODE ==="
  echo "Creating new task at: $TASK_DIR"
  mkdir -p "$TASK_DIR"
fi

### For NEW Tasks (Initializer Mode)

If task_list.md does NOT exist in the task directory:

SKILL_DIR="${CLAUDE_PLUGIN_ROOT}/skills/autonomous-skill"
TASK_DIR=".autonomous/$TASK_NAME"

# Read the initializer prompt template
INITIALIZER_PROMPT=$(cat "$SKILL_DIR/templates/initializer-prompt.md")

# Execute initializer session
claude -p "Task: $USER_TASK_DESCRIPTION
Task Directory: $TASK_DIR

$INITIALIZER_PROMPT" \\
  --output-format stream-json \\
  --max-turns 50 \\
  --append-system-prompt "You are the Initializer Agent. Create task_list.md and progress.md in $TASK_DIR directory."

### For CONTINUATION (Executor Mode)

If task_list.md EXISTS in the task directory:

SKILL_DIR="${CLAUDE_PLUGIN_ROOT}/skills/autonomous-skill"
TASK_DIR=".autonomous/$TASK_NAME"

# Read the executor prompt template
EXECUTOR_PROMPT=$(cat "$SKILL_DIR/templates/executor-prompt.md")

# Read current state
TASK_LIST=$(cat "$TASK_DIR/task_list.md")
PROGRESS=$(cat "$TASK_DIR/progress.md" 2>/dev/null || echo "No previous progress notes")

# Execute executor session
claude -p "Continue working on the task.
Task Directory: $TASK_DIR

Current task_list.md:
$TASK_LIST

Previous progress notes:
$PROGRESS

$EXECUTOR_PROMPT" \\
  --output-format stream-json \\
  --max-turns 100 \\
  --append-system-prompt "You are the Executor Agent. Complete tasks and update files in $TASK_DIR directory."

### Step 4: Auto-Continue Loop

After each session completes, check remaining tasks and auto-continue:

#!/bin/bash
TASK_DIR=".autonomous/$TASK_NAME"
AUTO_CONTINUE_DELAY=3
SESSION_NUM=1

while true; do
  echo ""
  echo "=========================================="
  echo "  SESSION $SESSION_NUM - Task: $TASK_NAME"
  echo "=========================================="

  # Run the appropriate agent
  # ... execute session ...

  # Check completion
  if [ -f "$TASK_DIR/task_list.md" ]; then
    TOTAL=$(grep -c '^\\- \\[' "$TASK_DIR/task_list.md" 2>/dev/null || echo "0")
    DONE=$(grep -c '^\\- \\[x\\]' "$TASK_DIR/task_list.md" 2>/dev/null || echo "0")

    echo ""
    echo "=== Progress: $DONE/$TOTAL tasks completed ==="

    if [ "$DONE" -eq "$TOTAL" ] && [ "$TOTAL" -gt 0 ]; then
      echo ""
      echo "All tasks completed! Exiting."
      break
    fi
  fi

  # Auto-continue with delay
  echo ""
  echo "Continuing in $AUTO_CONTINUE_DELAY seconds... (Press Ctrl+C to pause)"
  sleep $AUTO_CONTINUE_DELAY

  SESSION_NUM=$((SESSION_NUM + 1))
done

### Step 5: Report Progress

After execution, display a clear progress report:

==========================================
  SESSION COMPLETE - Task: build-rest-api
==========================================

Task Directory: .autonomous/build-rest-api/

Tasks completed this session:
- [x] Task 5: Implement user authentication
- [x] Task 6: Add login form validation

Overall Progress: 18/50 tasks (36%)

Next tasks:
- [ ] Task 7: Create password reset flow
- [ ] Task 8: Add session management

Continuing in 3 seconds... (Press Ctrl+C to pause)

### Example 1: Start New Task

User: Please use autonomous skill to build a REST API for a todo app

Response:
1. Generated task name: "build-rest-api-todo"
2. Created directory: .autonomous/build-rest-api-todo/
3. Running Initializer Agent...
4. Created task_list.md with 25 tasks
5. Progress: 3/25 completed
6. Auto-continuing in 3 seconds...

### Example 2: Continue Existing Task

User: Continue the autonomous task "build-rest-api-todo"

Response:
1. Found task: .autonomous/build-rest-api-todo/
2. Current progress: 15/25 tasks
3. Running Executor Agent...
4. Completed tasks 16-17
5. Progress: 17/25 completed
6. Auto-continuing in 3 seconds...

### Example 3: List All Tasks

# List all autonomous tasks
ls -la .autonomous/

# Show progress for specific task
cat .autonomous/build-rest-api-todo/task_list.md

### Key Files

For each task in .autonomous/<task-name>/:

FilePurposetask_list.mdMaster task list with checkbox progressprogress.mdSession-by-session progress notes

### Important Notes

Task Isolation: Each task has its own directory, no conflicts
Task Naming: Auto-generated from description (lowercase, hyphens)
Task List is Sacred: Never delete or modify descriptions, only mark [x]
One Task at a Time per Session: Focus on completing tasks thoroughly
Auto-Continue: Sessions auto-continue with 3s delay; Ctrl+C to pause

### Troubleshooting

IssueSolutionTask not foundCheck .autonomous/ for existing tasksMultiple tasksSpecify task name explicitlySession stuckCheck progress.md in task directoryNeed to restartDelete task directory and start fresh
## Trust
- Source: tencent
- Verification: Indexed source record
- Publisher: feiskyer
- Version: 0.1.0
## Source health
- Status: healthy
- Item download looks usable.
- Yavira can redirect you to the upstream package for this item.
- Health scope: item
- Reason: direct_download_ok
- Checked at: 2026-04-30T00:34:30.890Z
- Expires at: 2026-05-07T00:34:30.890Z
- Recommended action: Download for OpenClaw
## Links
- [Detail page](https://openagent3.xyz/skills/autonomous-skill)
- [Send to Agent page](https://openagent3.xyz/skills/autonomous-skill/agent)
- [JSON manifest](https://openagent3.xyz/skills/autonomous-skill/agent.json)
- [Markdown brief](https://openagent3.xyz/skills/autonomous-skill/agent.md)
- [Download page](https://openagent3.xyz/downloads/autonomous-skill)