菜单
Initialize or resume a long-running task session. Use when starting a complex multi-session task, resuming work from a previous session, or when the user mentions claude-progress.json or long-running work.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Trigger autonomous curiosity-driven exploration. The soul picks a topic from memory gaps or curiosity seeds, searches the web, and stores what it finds as dream-tagged memories.
Fine-tune the Qwen3-0.6B hint model — corpus gen, LoRA/unsloth, GGUF export, Ollama
Review soul discoveries (fixes, improvements, corrections) one by one, accept or discard each, implement accepted ones, build chitta, and optionally release.
First-principles review — question requirements, delete unnecessary parts, simplify, optimize with evidence, automate last. Use for code review, refactor, performance, or architecture.
Token-savvy session continuation. Rebuilds working context from transcript + soul memories in ~1500 tokens instead of replaying full history. Use when starting a new session to continue previous work.
Resume a thread by loading its ~800-token context capsule
| name | long-task |
| description | Initialize or resume a long-running task session. Use when starting a complex multi-session task, resuming work from a previous session, or when the user mentions claude-progress.json or long-running work. |
| execution | direct |
Mind-powered persistence for tasks that span multiple sessions.
/long-task - Show active task or start new one/long-task status - Show current task status/long-task complete - Mark task as completedWhen starting a long-running task:
Use long_task_start with:
- task_id: unique identifier (e.g., "implement-feature-x")
- goal: clear description of what you're trying to achieve
- hard_checks: deterministic completion criteria (optional)
- soft_checks: semantic completion criteria (optional)
long_task_event[TASK_COMPLETE] or call long_task_complete| Tool | Purpose |
|---|---|
long_task_start | Start a new long-running task |
long_task_get | Get task details by ID |
long_task_active | Get active task for realm |
long_task_update | Update progress (completed_summary, work_items, blockers) |
long_task_complete | Mark task as completed |
long_task_event | Log event (decision, error, checkpoint) |
long_task_snapshot | Get synthesized context |
long_task_evaluate | Check completion criteria |
long_task_activelong_task_snapshotlong_task_activelong_task_completeStart a task:
/long-task
> No active task. What's your goal?
> Implement the authentication system
Starting task: implement-auth
Goal: Implement the authentication system
Check status:
/long-task status
[LONG_TASK:implement-auth] Implement the authentication system
Iteration: 3 | Status: active
Done: Added login form, API endpoints
Pending: Add password reset; Add OAuth
Complete task:
/long-task complete
Outcome: Authentication system fully implemented with login, logout, and OAuth
Completed task: implement-auth
The Stop hook detects these markers to auto-complete:
[TASK_COMPLETE][DONE]<promise>COMPLETE</promise>Tasks support lease-based coordination:
task_claim: Claim task with leasetask_heartbeat: Extend lease