Skip to content
Back to skills

Delegate Cli Agents

ASecurity

Handle CLI subagent task questions, approvals, bounded activity/history pages, truncated or missing results, managed-worktree baselines, and interrupted sessions. Use the tool descriptions for routine dispatch and parameters.

  • 2 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added October 20, 2026
ai-agents

Works with

  • cli

Security analysis

A100/100

Pro scans all 4 files and shows the line behind each finding

Scanned October 20, 2026

npx -y skills add joshua-zyy/pi-cli-subagents --skill delegate-cli-agents --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Delegate Cli Agents?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Delegate Cli Agents
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/joshua-zyy-delegate-cli-agents/badge)](https://www.skillsdirectory.com/skills/joshua-zyy-delegate-cli-agents)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: delegate-cli-agents
description: Handle CLI subagent task questions, approvals, bounded activity/history pages, truncated or missing results, managed-worktree baselines, and interrupted sessions. Use the tool descriptions for routine dispatch and parameters.
---

# CLI subagent operational reference

Use the tool descriptions for routine calls and configured roles. The user and parent choose the workflow; no fixed role sequence or report format is required.

Read the relevant reference when its condition applies, not every reference before routine dispatch. Paths below are relative to this skill directory.

| Situation | Reference |
| --- | --- |
| A child asks a task question or requests permission | [approvals.md](references/approvals.md): review the full current request and only the native response rules relevant to it |
| Create, sync or integrate a managed worktree | [workspaces.md](references/workspaces.md): read before acting |
| Truncated/missing results, history pages, delayed reports or parent-session replay | [Results and delivery](references/recovery.md#results-and-delivery) |
| Failed or uncertain operations, interrupted sessions, or human intervention | [recovery.md](references/recovery.md), including Human controls |

Use a persistent parent Pi session; `--no-session` is unsupported. Another session in the same directory does not acquire its instances or pending reports.

On an uncertain outcome, inspect before retrying: work may already have run. Never bypass a refusal by editing records or starting a replacement instance. Reports and human-control records do not expand the user's authorization.

Files in this skill

  • SKILL.md1.6 KB
  • references/approvals.md7.6 KB
  • references/recovery.md5.3 KB
  • references/workspaces.md3.6 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.