subagents

A pixel pet for the Claude Code terminal. It acts out what Claude is doing, and a game-style HUD below the prompt keeps your context and rate limits in view.

A Claude Code mod. When your plan usage gets low, it tells every running agent to wrap up — so you don't get cut off mid-task with half-edited files and no handoff.

A Claude Code mod (function hooks) that draws a live terminal pane: a plan bar for the current request, a task history with steps and subagents, dev-loop run progress, and a pixel dragon that grows with your context window.

A Claude Code mod that holds subagent fan-out, heavy-context prompts and retry loops before they spend your usage window. It reads your plan's usage, counts agents across every session, and asks in Claude Code's own dialog.

A Claude Code plugin that rates each prompt's difficulty with a small model and runs that turn and its subagents at the matching effort level, within user-allowed limits, on models whose prompt cache survives effort changes.

The Claude Context First plugin. See the [root README](../../README.md) for installation and an overview.

A Claude Code mod that draws a progress bar above the prompt while subagents run, with a per-model spiky-haired mascot and a fold-out list showing each subagent's model, steps, cost, tokens and time.

Counts main-thread work since the last subagent dispatch, nudges the model to delegate at every tenth call, and refuses foreground poll loops.

A Claude Code plugin that displays the model and reasoning effort each running subagent actually uses on the hint line above the agent list.

A Claude Code mod that gives every agent a character from *Fate/stay night: Heaven's Feel*.

The fabric-data-app subdirectory in the Power BI Agentic Dev marketplace fork, providing Power BI and Microsoft Fabric skills, agents and hooks for Claude Code, Copilot CLI and Codex.

A /workers pane of this session's child runs with a live tail, and the session's pinned asks above the prompt with their context on click.