Subagents

Pi extension for single-agent delegation and scripted multi-agent workflows

Pi extension for durable background shell tasks, read-only delegated agents, local attested Pi runs, and fixed-purpose Fusion workflows through child Pi processes.

A pi extension that brings Claude Code-like sub-agents and workflow orchestration to pi — parallel execution, live widget, fleet view, custom agent types, mid-run steering, dynamic workflows, Claude Code compatibility, look and feel.

Claude-Code-style dynamic workflows for Pi — fan a task out across 100s of subagents with real model routing, token/cost accounting, resume, git-worktree isolation, an interactive /workflows TUI, and a real /deep-research.

Turn Pi into el Gentleman: a senior-architect development harness with SDD/OpenSpec, subagents, strict TDD evidence, review guardrails, and skill discovery.

Deterministic task planning and spec-orchestration for local models — crash-safe /task pipelines with verify/enforce gates, a real-time remote web view, and web/docs/fetch/worker subagent tools.

A focused, in-process sub-agent core for pi — autonomous agents plus a typed API and lifecycle events other extensions build on. Friendly fork of @tintinweb/pi-subagents.

Delegate bounded direct work or durable checked isolated task graphs to Pi roles.

Subagent orchestration for the pi coding agent — a thin dispatcher runs parallel subagents under a verification contract, keeping their output out of its context window. Multi-agent fleets, 15 personas, skills, herdr panes, peer coms, Hermes desktop.

Installable Pi package that adds markdown-defined subagents, delegated task tools, history, and model profiles.

Persistent, dependency-aware task graphs and subagent delegation for Pi coding sessions.

Deterministic multi-agent workflow orchestration for Pi

Workflow orchestration for Pi subagents.

Spec-gated, human-signed SDLC harness for pi (zense = a pun on the Thai word for sign) — sub-agents, dual eval, escalation gates — plus the Zense dark theme.

All-in-one multi-agent orchestration for the Pi coding agent — parallel teammate dispatch, goals, plans, knowledge system, MCP/LSP/browser, and cockpit visualization in a single install

Task workflow tracking for the pi coding agent: TODO/README maintenance, per-task git auto-commits as checkpoints, and an interactive /rollback. Hashline-aware, nested-repo aware, subagent-artifact-safe.

The yaag pi extension: run Orchestration Programs from a pi session.

Spawn scoped subagents from within a pi session: agent_run/agent_result tools, Claude Code-compatible agent types, turn caps and tool allowlists

Deterministic multi-step agent orchestration for pi: a Claude Code-style workflow tool with agent()/parallel()/pipeline() scripts over the shared agent catalog

Reliable Antigravity delegation for Pi — quota-aware runs, resumable tasks, repo verification, cancellation, and trustworthy diff summaries.

A Pi extension for configurable delegation intensity and exact subagent role model references.

A skill-based development workflow for Pi Agent. 30 contract-carrying skills (discover → research → design → plan → implement → validate → code-review → commit, and more), the 15 named subagents they dispatch, and four built-in /wf workflows that chain th

Role-based subagent orchestration for pi — delegates tasks to specialized pi child processes with configurable model roles

External Claude Code, Codex CLI, Antigravity, Grok Build CLI, Muse Code, and OpenCode delegation for pi.

Multi-role LLM orchestration via a guarded, observable handoff state machine — installs as a pi extension exposing `/conduct`.

In-process subagents for Pi with isolated SDK sessions, parallel and chained delegation, and inspectable threads.

Orchestrator pattern for pi: specialist subagents, enforcement, code review loop, workflow commands

Prime Agent integration for feature-factory: /feature command and Prime delegation adapter.

Shared multi-agent task board for Pi coding agent — Kimi Tower style orchestration (todo-flavored)

Swarm-first multi-agent messaging and task orchestration extension for Pi

Control-first, session-connected subagents for Pi with live navigation and intent-tier routing

Capability governance for pi sub-agents: spawn Agent Skills (SKILL.md) definitions whose allowed-tools becomes a grant that can only narrow going down a delegation tree, enforced by pi's own --tools allowlist, with an append-only ledger.

Passive Obsidian vault extension for pi. Watches @agent markers, dispatches forked subagents (Miner, Broadcaster, Heavy-Lifter), stores in LanceDB with vector + FTS + graph. Multi-agent 'Drop & Forget' workflow for the pi agent ecosystem.

Subagent extension for Pi coding agent. Delegate tasks to specialized agents.

Pi, enhanced — one install for nklisch's full harness: policy-gated command review, plugin marketplace, subagents, background tasks, research tools, search, model modes, and a curated UX set.

🐑 No-fuss herdr native subagent orchestration 🐑

Multi-phase security audits with specialist sub-agents, isolated context windows, capped concurrency, and resumable state — packaged as a Pi extension.

Observable multi-agent orchestration for Pi: autonomous subagents, isolated worktrees, swarms, schedules, handoffs, and a live operator dashboard.

Pi extension for fan-out code review: N parallel reviewer subagents + an inline gate via subagent workflowScript

Subagent dispatch with live TUI streaming and cost tracking

The armory suite's subagent orchestrator for the pi coding agent — a cross-harness, superpowers-native fleet where every agent is armory-native from birth.

pier: pi extension that fuses pi agent sessions with the herdr pane/tab orchestrator — todo_write loop closure + interactive subagents in isolated panes

Production-ready agentic SDLC framework for Pi and coding agents — orchestrator, builder/validator teams, lifecycle state, and specialist reuse

Reusable agent workflows and observable, attachable sub-agents for the Pi coding agent

Persistent nested subagents with bounded delegation for Pi

Delegate Pi work to focused child agents: code review, scouting, implementation, parallel audits, saved chains, and background jobs.

Delegate coding-agent work by spawning other agents as herdr tabs.

Integrated Mantice runtime for Pi: routed models, mechanical request reduction, accountable goals, managed subagents, and an event-driven activity sidebar

A repeatable Plan, Go, Close workflow for AI coding sessions: /session reads your repo and agrees the scope, /go runs the work in waves with a quality gate between each, /close verifies and commits. Runs on Claude Code, Codex CLI, Cursor and Pi.

Transient subagents for Pi: mini_subagent tool (single + parallel), /subagents command, and a live TUI panel

Concurrent worker orchestration for Pi

Thin Pi adapter for an explicit subagent provider and Rust Component-owned native lifecycle exchanges.

Capability-aware vision + paste extension for the pi coding agent. Delegates image analysis to a vision model only when the active primary model is text-only; passes images through natively for multimodal models (zero delegation).

Delegate tool for the Pi coding agent.

Run bounded, read-only Pi subagents without filling the parent context with intermediate investigation output

Unified agent and subagent management for pi coding agent, with OpenCode-compatible agent definitions.

Interactive async subagents for pi — spawn, orchestrate, and manage sub-agent sessions in multiplexer panes. Fork of HazAT/pi-interactive-subagents.

Pi extension for detached, sandboxed subagent runs that keep the foreground session free.

Adaptive delegation, cost-aware model routing, and deferred workflows across coding harnesses.

Bounded in-house subagents for Pi: child sessions read a frozen source snapshot and report back under fixed call, token and time ceilings

Pi agents that work together – three primitives (spawn/stop/send) build a tree of named, message-passing agents

Lightweight sub-agents for pi — spawn specialized agents with isolated sessions, tools, and models.

Pi extension, skill, and dependency-free Python CLI for coordinating coding agents in tmux

Delegate work to any harness (Claude Code, Muse, OpenCode, Amp) from the pi coding agent — code reviews, plans, implementation, security audits, docs, or your own custom templates.

Agent-agnostic CLI and catalog of skills, subagents, prompts, guardrails, and packs for safe Angular, TypeScript, and frontend engineering.

Delegating task/subagent extension for Pi: foreground/background subagents, widgets, tmux observability, SDK fallback.

Claude-Code-style dynamic workflow orchestration for Pi.

A generic framework for agent orchestration in pi.

pstack for Pi: rigorous agent workflows you can parallelize with confidence - poteto-mode playbooks, engineering principles, multi-model review panels, and subagents.

Maestria methodology for Prime Agent - specialist roles, orchestrator, global rules, and workflow modes as Agent Skills, plus a small Prime/Pi extension for mode commands and mode prompt injection

Pi extension that adds /review via an isolated review subagent.

Interactive multi-agent orchestration, messaging, and live sessions for the Pi Coding Agent

按语义结构读取 pi session 文件（对话历史）的工具：`session_read` 提供 11 个 action，覆盖 session 定位、fork/subagent/workflow 家族追踪、渐进式阅读（概览 → 单 turn → 全文）、全文/跨 session 搜索、素材提取、workflow run 概览、subagent 结果获取与环境自检——不需要直接读原始 `.jsonl`。

Deterministic JavaScript orchestration workflows backed by pi-subagents managed spawning.

pi extension: show assistant output token speed and TTFT in the footer, aggregated across in-process subagents

Focused multi-agent delegation with an interactive under-editor task list for Pi

Local Pi session analytics for token usage, cost, models, tools, subagents, skills, and history

Claude Code-style autonomous subagents and workflow orchestration for Pi, with HerbertGao-maintained UI extensions.

Claude-Code-style dynamic workflow orchestration for Pi. Fork of michaelliv/pi-dynamic-workflows.

A focused Pi package: web access, bounded subagents, visible todo tracking, computer use, LSP diagnostics, and health checks.

Pi 交互式 AI 代码审查扩展（支持单模型与多 Subagent 并发 2~6 专家自由切换，Codex 经典架构）

Pi extension that asks for model and thinking before every new pi-subagents run.

Google Antigravity CLI (agy) bridge for Pi — delegate bulk scaffolding, refactors, and test generation to a leaner model.

OpenPI — a Pi-native multi-agent workbench with background execution, isolated subagents, replay-safe workflows, goals, tasks, and observable TUI

Delegate focused work to context-aware Pi sub-agents

Durable, continuable subagents for Pi with spawn/fork providers and lifecycle controls.

Set goals in plan.md; a smart supervisor guides cheap worker subagents through long autonomous sessions until your goals are signed off, with every agent's pane visible to you.

Lean Pi subagents extension with full features: 268 initial tokens (96.5% fewer than current upstream).

Adaptive semantic-depth workflow orchestration for Pi, with focused, standard, and deep modes, isolated subagents, worktree delivery, structured output, and durable resume.

Pi extension that delegates autonomous implementation tasks to Google's official Antigravity ACP server.

Cognitive-driven multi-agent orchestration across Agent Kernel, Prompt Vault, ROCS, and Pi runtime owners

Recursive, isolated, asynchronous subagents for Pi

Subagent extension for Pi coding agent. Delegate prompts to specialized agents with optional named sessions.

An ARC-focused Pi distribution with bounded multi-model delegation.

Explicit local communication between Pi sessions; Pi orchestrates.

Blocking + background subagents for skills that require subagents (code-review, design-it-twice, improve-codebase-architecture, research, wayfinder, grilling).

A maintained MIT fork of pi-subagents with ordered lifecycle interception for Pi extensions.

ACP adapter for pi coding agent (fork), with a bundled pi extension for subagent→plan

pi extension that records agent work time per prompt, excluding waits for the user, with subagent linkage and task/session views

Claude Code-style autonomous sub-agents for the Pi coding agent, with a step-summary conversation viewer.

Persistent foreground and background subagents for pi

Online self-improvement layer for pi. A unified, in-trajectory harness-state store (prompt notes, memory, skill descriptions, sub-agent specs) with a manual /refine that proposes evidence-backed structured CRUD deltas. Composes with pi-reflect (offline) a

Minimal subagent tool extension for pi-agent: delegate a task to a fresh headless pi instance.

Four-line omp statusline widget: zai coding-plan quota (provider-gated), prayer times, hijri clock, and API-spend ledger (REPO/DAY/7DAY/30DAY, subagent-inclusive).

Non-blocking subagent orchestration for pi coding agent

Lets extensions' interactive dialogs from nested subagents render in the root session, with the subagent's last message as context, via simple integration.

Run bounded isolated Pi subagents.

Delegate pi work to isolated, budgeted children with verification loops and tracing.

Pi extension for spawning sub-agents via RPC for parallel task execution

Delegate tasks to specialized subagents with isolated context windows.

Pi sub-agent orchestration package for Yishan.

Subagent extension for pi with configurable model and thinking mode via selection UI

A subagent tool supporting context compaction and nested subagents — user-customizable models (with round-robin) and tool policies, extensible via hooks.

sling extension for delegating tasks to subagents with chains, parallel execution, and TUI clarification

Local-first memory platform and orchestration for AI coding agents

Model-facing front door to the first-party workflow engine — run JS workflow scripts over the subagent runtime, replacing the third-party @quintinshaw/pi-dynamic-workflows extension

Pi Agent extension that adds a describe_image tool, letting non-multimodal models delegate image analysis to a vision-capable model (like Qwen VL)

Pi extension for delegating work to project or user agents, including persistent tmux agent panes.

Stage-gated pre-production orchestration for Pi — Brainstorm → PRD → RTM → Feasibility → Design → Pseudocode → Test Plan, with 4-scout visible subagents per stage and structured doctor audit.

Agent orchestration for Pi

Protocol adapter that lets @tintinweb/pi-tasks TaskExecute spawn nicobailon/pi-subagents.

Stronger answers for hard Pi questions via a parallel model panel + judge, built on pi-subagents

First-party subagent dispatch and fleet: fan out parallel child agents and inspect their runs — no pi-subagents dependency

Subagent delegation for pi: single, parallel, and chained runs in isolated pi processes, with a user-defined agent roster, model resolution, and slash commands.

A dialog coordinator for pi extensions that prevents dialogs from rendering over each other, plus a select-with-note component that wires subagent routing and notification.

A compact Pi footer showing model, Git, context, quotas, and active work/subagent status.

Sub-agents extension for pi coding agent.

Git worktree isolation for @gotgenes/pi-subagents — a WorkspaceProvider that runs subagents in isolated worktrees.

A lightweight subagent spawning and managing system for pi.

Durable, low-noise background subagents for Pi

Codex-shaped, session-scoped subagents for Pi with templates, waits, steering, and a live overlay.

Pi extension. Delegate a batch of read-only research tasks to child agents, with dependency edges between them, a model and thinking effort you pick, and a two-way channel back to the parent.

Pi runtime telemetry: compaction, tools, vault, skills, follow-ups, subagents -> NDJSON shards, HTML dashboard, agent tool

Interactive subagents for pi running in herdr

A Pi package extension that provides sub-agent functionality.

SAP AI Core (orchestration + foundation) provider for the pi coding agent

An opinionated Pi workflow suite for planning, intent clarification, permissions, subagents, Git automation, project initialization, and publishing.

Claude Code-style dynamic workflow orchestration for Pi, built on @agwab/pi-workflow.

A question tool for the agent — single/multi-select or free-text, with subagent forwarding and attention alerts.

Pi extension that runs Claude Code headlessly through an agent-safe CLI runner with a no-silent-hang guarantee (wall-clock + stall watchdogs, process-group kill, guaranteed completion hook), injects a completion hook back into Pi, and bundles a delegation

Create sub-agents dynamically and run them in parallel; single blocking tool, no orchestration, nesting prevented

Asynchronous Pi subagents in Herdr, with optional isolated Git worktrees (fork: worktree creation aligned with wt)

Prompt-native isolated Pi subprocess execution for single, parallel, chain, and workflow runs

Task-first child delegation for Pi: foreground runs, background completion notices, and bounded subprocess lifecycle.

Delegate tasks to subagents with isolated context, choosing the model and thinking level per call.

Run any shell command and pi subagents under babysit, with context-safe captured output.

Supervisor-led multi-Agent orchestration for Pi with durable workflows, quality gates, and observable subagents

Lightweight sub-agents for pi — spawn specialized agents with isolated sessions, tools, and models.

Persistent dynamic task orchestration for pi

Pi tool that delegates a self-contained sub-task to OpenAI's Codex CLI via `codex exec --json`. Streams structured events as partial output, returns the final agent message. Configurable default model + reasoning effort, with multi-turn session continuity

A modular Pi extension for terminal UI, session workflows, adversarial review, and delegated agents

Background subagents on a pi or Claude Code backend, with fire-and-forget spawn and deferred result delivery.

Single pi package for the entire /plan grill-and-plan workflow — one interactive planner does grilling + domain-modeling → pointer plan + todos → sequential workers → reviewer, all on herdr subagents.

Multi-model and solo coding orchestration for Claude Code, Codex, and pi.

Multi-agent orchestration across pi sessions

Kimi Code-style agent orchestration harness for Pi coding agent — Swarm (concurrent subagents + braille TUI), Goal (lifecycle + budget + queue), Plan (plan mode + tool restrictions), Permission (18-level policy chain: auto/yolo/manual), Task (background +

Live tokens-per-second (TPS) meter for Pi with gentle-pi subagent visibility

Sandboxed multi-agent workflow orchestration, plus a deterministic multi-angle code review built on it.

Subagent tool for pi: delegate tasks to specialized agents with isolated context windows (single, parallel, chain modes) with streaming output, plan-mode propagation and parent-model inheritance

Reusable Pi capability profiles plus context-isolated delegation and pluggable child surfaces

将复杂任务拆成任务图并交给隔离 Pi 子进程执行的 Pi 扩展

Subagent tool for Pi: one task per call, inline or named agent personas, resumable sessions, background runs with subagent_wait, a live subagent panel, and a parent-side approval coordinator for the permission gate extension.

Pi tools and package support for subagent orchestration.

Live Pi session and subagent observability for Linux desktops

Lost the thread? Get a short briefing on your Pi session and subagents, then ask questions in the same TUI panel. Uses your active model.

Adds the agent_team delegation tool and pi-multiagent skill to Pi.