ClaudeMods
☰
EN
● 0 online · Views 0 times
SponsorsSubmit a project
GitHub repositories · by wolf123450

subagent-tree

A hooks-module Claude Code mod that adds an "Agents" side pane showing subagents in a tree with last-tool activity, tokens and runtime per agent, click-through detail, search, and saved session history.

wolf123450@wolf123450

wolf123450/ClaudeSkills/tree/master/plugins/subagent-tree

Translated

About this mod

subagent-tree is a hooks-module mod distributed through the wolf123450/ClaudeSkills plugin marketplace. It adds an "Agents" side pane to Claude Code: subagents are shown in a tree with the last tool used and how long ago it ran, plus tokens and runtime per agent. Clicking an agent reveals detail (prompt, tool log, result); the pane also supports search and saved history of past sessions. Install with /plugin marketplace add wolf123450/ClaudeSkills followed by /plugin install subagent-tree@claude-skills.

Installation

Check the author's README for the marketplace and plugin name first. Commands may change as the repository evolves.

claude plugin marketplace add wolf123450/ClaudeSkills
claude plugin install subagent-tree
Original text / README

ClaudeSkills

Clint's personal Claude Code plugin marketplace — skills and agents distributed as installable, auto-updating plugins instead of hand-zipped .skill files.

Plugins

  • dev-workflow — Playwright testing tips (playwright-testing skill), a playwright-runner subagent that executes Playwright tests/browser steps and reports back a compact summary, and a demoing-work skill for walking the user through completed work and gating "done" on their explicit confirmation.
  • game-catchup — Generates a structured "what you missed" recap for a game you haven't played in a while.
  • feature-maturity — Shared vocabulary for how done a feature actually is: a Completeness ladder (Spike/Prototype/Functional/Finished) plus independent quality axes (Polish, Robustness, Architecture, Verification, Documentation, Accessibility). Use for scoping work, self-reporting before declaring something "done," auditing existing code, or resolving disagreements about whether something is really Finished/Polished/Hardened.
  • desktop-statusline — A hooks-module mod that mirrors the terminal status line (model, session tokens, context %, 5h/7d rate limits with reset countdown) as a band above the prompt in the Claude desktop app's Code tab. Hidden in the terminal, where the real status line runs.
  • subagent-tree — A hooks-module mod that adds an "Agents" side pane: subagents in a tree with the last tool and how long ago it ran, tokens and runtime per agent, click-through detail (prompt, tool log, result), search, and saved history of past sessions.

Install

/plugin marketplace add wolf123450/ClaudeSkills
/plugin install dev-workflow@claude-skills
/plugin install game-catchup@claude-skills
/plugin install feature-maturity@claude-skills
/plugin install desktop-statusline@claude-skills
/plugin install subagent-tree@claude-skills

(Or via the claude CLI: claude plugin marketplace add wolf123450/ClaudeSkills, then claude plugin install dev-workflow@claude-skills.)

Updating

/plugin marketplace update claude-skills

Or just wait — Claude Code periodically checks configured marketplaces for updates. Since plugins live in this git repo, any commit pushed here is what the next update pulls; no manual re-zipping or redistribution step.

Adding a new plugin

  1. Create plugins/<name>/.claude-plugin/plugin.json (name, description, version, author).
  2. Add its skills under plugins/<name>/skills/<skill-name>/SKILL.md and/or agents under plugins/<name>/agents/<agent-name>.md.
  3. Add an entry to .claude-plugin/marketplace.json's plugins array with "source": "./plugins/<name>".
  4. Run claude plugin validate . from the repo root to confirm it's picked up cleanly before committing.

Similar projects