Requirements
- Target platform
- OpenClaw
- Install method
- Manual import
- Extraction
- Extract archive
- Prerequisites
- OpenClaw
- Primary doc
- SKILL.md
Tencent SkillHub · AI
Task Engine
Multi-agent task orchestration engine with state machine tracking. Use when complex multi-step projects need automated monitoring, multi-agent collaboration,...
skill openclawclawhub Free
Multi-agent task orchestration engine with state machine tracking. Use when complex multi-step projects need automated monitoring, multi-agent collaboration,...
⬇ 0 downloads ★ 0 stars Unverified but indexed
Install for OpenClaw
Item is unstable.
This item is timing out or returning errors right now. Review the source page and try again later.
Quick setup
- Wait for the source to recover or retry later.
- Review SKILL.md only after the source returns a real package.
- Do not rely on this source for automated install yet.
Package facts
- Download mode
- Manual review
- Package format
- ZIP package
- Source platform
- Tencent SkillHub
- What's included
- CLAUDE.md, DESIGN.md, SKILL.md, config/settings.yaml, references/agent-capabilities.md, references/state-transitions.md
Validation
- Wait for the source to recover or retry later.
- Review SKILL.md only after the download returns a real package.
- Treat this source as transient until the upstream errors clear.
Install with your agent
Agent handoff
Use the source page and any available docs to guide the install because the item is currently unstable or timing out.
- Open the source page via Review source status.
- If you can obtain the package, extract it into a folder your agent can access.
- Paste one of the prompts below and point your agent at the source page and extracted files.
New install
I tried to install a skill package from Yavira, but the item is currently unstable or timing out. Inspect the source page and any extracted docs, then tell me what you can confirm and any manual steps still required.
Upgrade existing
I tried to upgrade a skill package from Yavira, but the item is currently unstable or timing out. Compare the source page and any extracted docs with my current installation, then summarize what changed and what manual follow-up I still need.
Trust & source
Release facts
- Source
- Tencent SkillHub
- Verification
- Indexed source record
- Version
- 1.0.0
Provenance
- Publisher
- zeron-G
- Source page
- View original listing
- Canonical URL
- Open canonical page
Documentation
Task Engine
Orchestrates multi-step projects across multiple agents (Claude Code, Eva, others) using a state machine with JSON-based persistence. Each task has subtasks dispatched to agents, tracked via heartbeat, and reported through Discord.
CLI Commands
All commands run from the skill root: cd /home/zeron/.openclaw/workspace/skills/task-engine All commands support --json for machine-readable output: {"ok": true, "task_id": "TASK-001", "status": "PLANNING", "message": "Created TASK-001"}
Create a task
python3 scripts/task_engine.py create "Implement feature X" --priority P0 --plan "3 phases: models, API, tests" python3 scripts/task_engine.py create "Feature Y" --priority P1 --json
View status
python3 scripts/task_engine.py status # All active tasks (table) python3 scripts/task_engine.py status TASK-001 # Detailed single task python3 scripts/task_engine.py status TASK-001 --json # Machine-readable python3 scripts/task_engine.py status --all # Include terminal tasks
Transition task state
python3 scripts/task_engine.py transition TASK-001 approve --note "Plan approved" --json python3 scripts/task_engine.py transition TASK-001 block --note "Waiting on API key" python3 scripts/task_engine.py transition TASK-001 complete --note "All verified" --json
Dispatch subtask to an agent
python3 scripts/task_engine.py dispatch TASK-001 "Implement auth models" \ --agent claude-code --type dev --json python3 scripts/task_engine.py dispatch TASK-001 "Run integration tests" \ --agent eva --type test --deps subtask_01,subtask_02 Dispatching the first subtask auto-transitions APPROVED -> IN_PROGRESS.
Update subtask progress
python3 scripts/task_engine.py subtask TASK-001 subtask_01 start --progress 30 --json python3 scripts/task_engine.py subtask TASK-001 subtask_01 done --note "Models complete" --json python3 scripts/task_engine.py subtask TASK-001 subtask_02 fail --note "Schema mismatch" --json
Check tasks (heartbeat integration)
python3 scripts/task_engine.py check # Check all active tasks (verbose) python3 scripts/task_engine.py check TASK-001 # Check one task python3 scripts/task_engine.py check --quiet # Minimal output for cron python3 scripts/task_engine.py check --json # Machine-readable JSON python3 scripts/task_engine.py check --discord # Discord-formatted digest
Archive completed task
python3 scripts/task_engine.py archive TASK-001 --json # Only works on terminal tasks
Auto-dispatch
Auto-dispatch scans subtasks and dispatches ready ones to appropriate agents: python3 scripts/task_engine.py auto-dispatch TASK-001 # Dispatch ready subtasks python3 scripts/task_engine.py auto-dispatch TASK-001 --dry-run # Preview without acting python3 scripts/task_engine.py auto-dispatch --all # All active tasks python3 scripts/task_engine.py auto-dispatch TASK-001 --subtask subtask_02 # Specific subtask python3 scripts/task_engine.py auto-dispatch TASK-001 --subtask subtask_01 --show-context # View dispatch context Output is always JSON with dispatches and skipped arrays.
Notify (Discord formatting)
Generate Discord-formatted notification messages: python3 scripts/task_engine.py notify digest # Full heartbeat digest python3 scripts/task_engine.py notify TASK-001 created # Task creation message python3 scripts/task_engine.py notify TASK-001 status # Status update with progress python3 scripts/task_engine.py notify TASK-001 transition # Last transition python3 scripts/task_engine.py notify TASK-001 completed # Completion summary python3 scripts/task_engine.py notify TASK-001 alert --type stuck --subtask-id subtask_01
Rebuild index (recovery)
Reconstruct index.json from task directories if it gets corrupted: python3 scripts/task_engine.py rebuild-index # Scan and rebuild python3 scripts/task_engine.py rebuild-index --json # Machine-readable output
Heartbeat Integration
Add step 4.3 to the heartbeat's cmd_beat() function, after the ongoing.json check: # 4.3 Task Engine check logger.info("[4.3/8] Task Engine check") try: import sys sys.path.insert(0, str(Path("/home/zeron/.openclaw/workspace/skills/task-engine/scripts"))) from engine.checker import check_all_tasks te_result = check_all_tasks() if te_result.get("alerts"): alerts.extend(te_result["alerts"]) all_ok = False if te_result.get("summary"): logger.info(" Tasks: %s", te_result["summary"]) except ImportError: logger.debug(" Task engine not installed, skipping") except Exception as e: logger.warning(" Task engine check failed: %s", e) The check is cheap (~300-500 tokens per heartbeat for 1-3 active tasks). It reads index.json first, then only loads task/subtask files for active tasks.
State Machine
PLANNING ──approve──> APPROVED ──start──> IN_PROGRESS ──test──> TESTING ──review──> REVIEW ──complete──> COMPLETED │ │ │ │ │ │ │reject block│ │fail reopen│ │fail reopen│ fail v v v v v v v REJECTED BLOCKED FAILED IN_PROGRESS FAILED IN_PROGRESS FAILED │ unblock│ v IN_PROGRESS Terminal states: COMPLETED, FAILED, REJECTED. See references/state-transitions.md for the full transition table.
Subtask states
PENDING -> ASSIGNED -> IN_PROGRESS -> DONE │ │ │block │fail / block v v BLOCKED FAILED / BLOCKED
Auto-transitions (checked by heartbeat)
ConditionActionAll type: dev subtasks DONETask -> TESTINGAll type: test subtasks DONETask -> REVIEWFirst subtask dispatchedTask APPROVED -> IN_PROGRESSAny subtask FAILEDAlert for human interventionSubtask stuck 3+ heartbeatsAlert via DiscordTask past ETAAlert as overdue
Discord Formatting
The discord_formatter.py module generates formatted messages for Discord notifications. All formatting is pure string generation — no API calls. Available formats: Task created: New task announcement with priority and plan Status update: Progress bars and subtask tree Transition: State change with emoji indicators Alert: Urgent stuck/overdue/failed alerts with human ping Completion summary: Final report with subtask results Heartbeat digest: Full summary of all active tasks Progress bars render as: [████░░░░░░] 40%
Agent Capabilities
AgentKeyBest ForMax ParallelClaude Codeclaude-codeDev, refactor, debug, docs3EvaevaTest, validate, system-ops1 Agent selection priority: Preferred agent (if specified and capable) Match by preferred_types Match by broader capabilities Fallback to Eva See references/agent-capabilities.md for full details.
Corrupted index.json
python3 scripts/task_engine.py rebuild-index --json Scans all tasks/TASK-*/task.json files and reconstructs the index. Skips invalid/unreadable files.
Common errors
ErrorCauseFixInvalid transition: X + 'event'Transition not allowed from current stateCheck references/state-transitions.mdTask TASK-XXX not foundTask doesn't exist or was archivedCheck tasks/ and tasks/archive/Cannot archive: not terminalTask must be COMPLETED/FAILED/REJECTEDTransition to terminal state firstAgent at capacityMax parallel instances reachedWait for running subtasks to completeDependency not DONEBlocked by incomplete subtaskComplete blocking subtask first
JSON parse errors
The engine handles corrupt JSON files gracefully: During check: skips bad tasks, logs warning, continues During rebuild-index: skips unreadable files, reports in output All CLI commands catch exceptions and return clean error messages (not tracebacks)
Data Location
Task files: /home/zeron/.openclaw/workspace/tasks/TASK-NNN/ Index: /home/zeron/.openclaw/workspace/tasks/index.json Archive: /home/zeron/.openclaw/workspace/tasks/archive/ Config: /home/zeron/.openclaw/workspace/skills/task-engine/config/settings.yaml Agent reference: /home/zeron/.openclaw/workspace/skills/task-engine/references/agent-capabilities.md State transitions: /home/zeron/.openclaw/workspace/skills/task-engine/references/state-transitions.md
Category context
Agent frameworks, memory systems, reasoning layers, and model-native orchestration.
Source: Tencent SkillHub
Largest current source with strong distribution and engagement signals.
Package contents
Included in package
5 Docs1 Config
- SKILL.md
- CLAUDE.md
- DESIGN.md
- references/agent-capabilities.md
- references/state-transitions.md
- config/settings.yaml