--- name: hzl description: Persistent task ledger for agent coordination. Plan multi-step work, checkpoint progress across session boundaries, and coordinate across multiple agents with project pool routing. metadata: { "openclaw": { "emoji": "🧾", "homepage": "https://github.com/tmchow/hzl", "requires": { "bins": ["hzl"] }, "install": [ { "id": "brew", "kind": "brew", "package": "hzl", "bins": ["hzl"], "label": "Install HZL (Homebrew)" }, { "id": "node", "kind": "node", "package": "hzl-cli", "bins": ["hzl"], "label": "Install HZL (npm)" } ] } } ---

HZL: Persistent task ledger for agents

HZL (https://hzl-tasks.com) is a local-first task ledger that agents use to:

• Plan multi-step work into projects + tasks
• Checkpoint progress so work survives session boundaries
• Route work to the right agent via project pools
• Coordinate across multiple agents with leases and dependencies

This skill teaches an agent how to use the hzl CLI.

When to use HZL

OpenClaw has no native task tracking. Unlike Claude Code (which has TodoWrite) or Codex (which has update_plan), OpenClaw relies on memory and markdown files for tracking work. HZL fills this gap.

Use HZL for:

• Multi-step projects with real sequencing or handoffs
• Work that may outlive this session or involve multiple agents
• Anything where "resume exactly where we left off" matters
• Delegating work to another agent and needing recovery if they fail

Skip HZL for:

• Truly trivial one-step tasks you will complete immediately
• Time-based reminders (use OpenClaw Cron instead)
• Longform notes or knowledge capture (use memory files)

Rule of thumb: If you feel tempted to make a multi-step plan, or there is any chance you will not finish in this session, use HZL.

---

⚠️ DESTRUCTIVE COMMANDS — READ FIRST

| Command | Effect | |---------|--------| | hzl init --force | DELETES ALL DATA. Prompts for confirmation. | | hzl init --force --yes | DELETES ALL DATA WITHOUT CONFIRMATION. | | hzl task prune ... --yes | PERMANENTLY DELETES old done/archived tasks and history. |

Never run these unless the user explicitly asks you to delete data. There is no undo.

---

Core concepts

• Project: container for tasks. In single-agent setups, use one shared project. In multi-agent setups, use one project per agent role (pool routing).
• Task: top-level work item. Use parent tasks for multi-step initiatives.
• Subtask: breakdown of a task (--parent ). Max 1 level of nesting. Parent tasks are never returned by hzl task claim --next.
• Checkpoint: short progress snapshot for session recovery.
• Lease: time-limited claim that enables stuck detection in multi-agent flows.

---

Project setup

Single-agent setup

Use one shared project. Requests and initiatives become parent tasks, not new projects.

hzl project list                    # Check first — only create if missing
hzl project create openclaw

Everything goes into openclaw. hzl task claim --next -P openclaw always works.

Multi-agent setup (pool routing)

Use one project per agent role. Tasks assigned to a project (not a specific agent) can be claimed by any agent monitoring that pool. This is the correct pattern when a role may scale to multiple agents.

hzl project create research
hzl project create writing
hzl project create coding
hzl project create marketing
hzl project create coordination    # for cross-agent work

Pool routing rule: assign tasks to a project without --agent. Any eligible agent claims with --next.

# Assigning work to the research pool (no --agent)
hzl task add "Research competitor pricing" -P research -s ready

# Kenji (or any researcher) claims it
hzl task claim --next -P research --agent kenji

Agent routing: when --agent is set at task creation, only that agent (or agents with no assignment) can claim it via --next. Tasks with no agent are available to everyone.

# Pre-route a task to a specific agent
hzl task add "Review Clara's PR" -P coding -s ready --agent kenji

# Kenji claims it (matches agent)
hzl task claim --next -P coding --agent kenji   # ✓ returns it

# Ada tries — skipped because it's assigned to kenji
hzl task claim --next -P coding --agent ada     # ✗ skips it

Use --agent on task creation when you specifically want one person. Omit it when any eligible agent in the pool should pick it up.

---

Session start (primary workflow)

With workflow commands (HZL v2+)

hzl workflow run start --agent <agent-id> --project <project> --json

--project is required — agents must scope to their assigned pool. Use --any-project to intentionally scan all projects (e.g. coordination agents).

This handles expired-lease recovery and new-task claiming in one command. If a task is returned, work on it. If nothing is returned, the queue is empty. Agent routing applies: tasks assigned to other agents are skipped.

Without workflow commands (fallback)

hzl agent status                           # Who's active? What's running?
hzl task list -P <project> --available     # What's ready?
hzl task stuck                             # Any expired leases?
hzl task stuck --stale                     # Also check for stale tasks (no checkpoints)

# If stuck tasks exist, read their state before claiming
hzl task show <stuck-id> --view standard --json
hzl task steal <stuck-id> --if-expired --agent <agent-id> --lease 30
hzl task show <stuck-id> --view standard --json | jq '.checkpoints[-1]'

# Otherwise claim next available
hzl task claim --next -P <project> --agent <agent-id>

---

Core workflows

Adding work

hzl task add "Feature X" -P openclaw -s ready              # Single-agent
hzl task add "Research topic Y" -P research -s ready        # Pool-routed (multi-agent)
hzl task add "Subtask A" --parent <id>                      # Subtask
hzl task add "Subtask B" --parent <id> --depends-on <a-id>  # With dependency

Working a task

hzl task claim <id>                          # Claim specific task
hzl task claim --next -P <project>           # Claim next available
hzl task checkpoint <id> "milestone X"       # Checkpoint progress
hzl task complete <id>                       # Finish

Status transitions

hzl task set-status <id> ready               # Make claimable
hzl task set-status <id> backlog             # Move back to planning
hzl task block <id> --comment "reason"       # Block with reason
hzl task unblock <id>                        # Unblock

Statuses: backlog → ready → in_progress → done (or blocked)

Finishing subtasks

hzl task complete <subtask-id>
hzl task show <parent-id> --view summary --json   # Any subtasks remaining?
hzl task complete <parent-id>               # Complete parent if all done

---

Delegating and handing off work

Workflow commands (HZL v2+)

# Hand off to another agent or pool — complete current, create follow-on atomically
hzl workflow run handoff \
  --from <task-id> \
  --title "<new task title>" \
  --project <pool>              # --agent if specific person; --project for pool

# Delegate a subtask — creates dependency edge by default
hzl workflow run delegate \
  --from <task-id> \
  --title "<delegated task>" \
  --project <pool> \
  --pause-parent                # Block parent until delegated task is done

--agent and --project guardrail: at least one is required on handoff. Omitting --agent creates a pool-routed task; --project is then required to define which pool.

Manual delegation (fallback)

hzl task add "<delegated title>" -P <pool> -s ready --depends-on <parent-id>
hzl task checkpoint <parent-id> "Delegated X to <pool> pool. Waiting on <task-id>."
hzl task block <parent-id> --comment "Waiting for <delegated-task-id>"

---

Dependencies

...