Untested on omp

Sidebar TUI extension for pi coding agent

Persistent nested subagents with bounded delegation for Pi

Declarative, user-definable permission modes for the pi coding agent: OS-level sandboxing (bubblewrap / sandbox-exec), an allow/ask/deny policy engine, real bash AST gating (tree-sitter), tool hiding, and skill/custom-tool gating.

Pi extension for switching named provider-owned OAuth accounts.

Route pi prompts to task-appropriate model tiers with TypeSafe Jev typed judgments. Budget-aware, with automatic fallback.

定时任务调度扩展：按 duration（`5m` / `2h` / `1d`）间隔或 cron 表达式，在指定时间向 agent 注入消息。支持一次性提醒（once）与过期策略（expires）。创建入口两路：**人侧 `/schedule` 命令打开创建表单**（表单异步打开，填表时长不受命令通道超时约束），**模型侧 `schedule` tool 直建**（不再弹确认表单；参数不完整时必须先向用户澄清——见「创建方式」）。任务随 owner session 持久化，resume 后继续触发。

Governed long-term memory, built-in session search, and optional Obsidian vault integration 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`。

Anti-slop UI/UX design discipline for your Pi agent — anchors a lintable DESIGN.md, derives a design direction (mood, type voice, color mood, signature), runs deterministic slop-audit gates (APCA contrast + tokens + states + slop tells), works with text-o

Records managed pi sessions for Agent Switchboard (swb)

Performance optimization suite for Pi Coding Agent - caveman mode + RTK tool rewriting + ponytail lazy-dev mode

Local-first RAG knowledge base for Pi and OMP coding agents with hybrid code/doc search, reranking, diagnostics, and large-project indexing

Pi coding agent extension for LSP (Language Server Protocol) integration

Claude, Codex, Meta Muse, OpenCode Go, and GitHub Copilot usage reports and a widget for Pi Coding Agent

Pi extension: search session history in a two-pane picker with full conversation preview, then resume or fork.

Shared helpers for pi extensions: the hook log envelope, TUI tool headers, the system-prompt block seam, and the session-to-session transport.

Posthorse: fresh context, same journey. No-summary context rollover for Pi, with sparse budget reminders, handoffs, notes, and history.

Evidence-first structural review of code changes, powered by Fovea's heat-diffusion substrate.

MCP tools for Pi, discovered on demand and called natively through the official SDK.

A Jev classification layer for the Pi coding agent: typed routing, permission gating, context shielding, pruning and loop detection.

Enhanced UX for managing local Pi extensions and community packages

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

Permission enforcement extension for the Pi coding agent. Fork of @gotgenes/pi-permission-system (v40.1.2) with configurable wrapper-floor behavior.

A live visual surface for terminal coding agents — agents draw HTML snippets, you watch them in the browser and comment back.

基于 AST 的 Pi 代理安全防护扩展 — 解析 Bash 命令，拦截危险操作（继承自 pi-damage-control）

CodeMem extension for the pi coding agent — ingest, injection, and native memory tools

Pi extension for live token usage monitoring and budget tracking (Day / Week / Month / Total). 0 prompt tokens.

pi extension: named modes bundling instructions, tool set, model, subagent model and colour — create/enable/disable with /mode, switch with ctrl+tab.

Huddo for pi: join group chats where people and AI agents talk together. Adds the Huddo MCP server and the huddo skill.

Pi coding agent extension - henyo core functionality

Native, explainable automatic model routing for the Pi coding agent

Shepherd pi skills and role resources.

Session-aware todo list overlay for Pi: todo tool, /todos command, and a live TUI panel

Pi package for configurable, approved, resumable task workflows.

Interactive Terminals, Background jobs, and a /ps panel for Pi.

Watchdog for the pi coding agent: detects bash commands that produce no output (hung builds, silently stuck dev servers) and warns you instead of waiting forever.

AI agent orchestration for pi — parallel task execution with checkpoint discipline

TopoRealm 1.0 — single-owner daemon graph workspace (aggregate: toporealm CLI + toporeald daemon). Minimal core, two-layer modules, one local graph shared by humans and agents.

Jev (TypeSafe System One) backed auto mode for the Pi coding agent: semantically auto-approves bash, write, and edit tool calls and fails closed when a decision cannot be made.

Kiro API-key provider for pi using KIRO_API_KEY. Vendors pi-kiro's streaming core (MIT) with no runtime dependencies.

Codex-style $skill-name invocation anywhere in a pi prompt

Typed lifecycle, protocol, and Leader contracts for independently bundled DoomPi extensions.

Experimental Codex Sky desktop tools: native MCP for Claude Code, plus pi native/optional Jev modes

Shared library for Pi usage monitoring extensions

A generic pi provider extension for OpenAI-compatible, Anthropic-compatible, OpenAI Responses, and Ollama chat endpoints

Minimal Pi extension that automatically compacts context at a configurable usage threshold.

Per-git-branch .pi workspace context for pi

MCP (Model Context Protocol) adapter extension for Pi coding agent

Voice for Pi coding agents — /speak reads each turn's final answer aloud, /talk records and transcribes voice input into the composer, both through pluggable providers (local OS voices, any OpenAI-compatible endpoint, or a host-registered provider such as

Pi-Pi: a persistent, general-purpose Pi agent with adaptive tools, workers, skills, and memory