Untested on omp

GLM Coding Plan (Zhipu BigModel credit-based plans, incl. team) quota usage in the pi footer — org/project context headers, /glm-quota report, /glm-quota-setup wizard

Claude Code–style startup header for Pi — animated Pi logo with a 9-color palette, toggleable IBM stripes and Minecraft themes; configurable info bar displays version, model, resource stats, AGENTS.md marker, cwd, and a customizable slogan.

pi extension (prompt-level): enforce ffgrep/fffind-first search discipline and neutralize the default bash file-operation guideline

Generate reviewed Git commits with a dedicated configurable model in the pi coding agent.

Miao Vision extension and skill for the Pi coding agent

breadcrumbs an agent leaves while working with you, kept as records in a space on your PDS. a pi extension.

Pi package containing Yishan's development workflow skills: ideation, planning, debugging, testing, review, subagent orchestration, and whiteboard (Excalidraw scene) editing.

Configurable status bar extension for the pi coding agent.

Full Telegram control of pi coding agent — commands, menus, interactive UI, model management, and more

Fork the current Pi session into a split terminal pane or new tab with rotating layout policies, and with optional model and message queuing, and support for cmux, tmux, iTerm2, Terminal.app, Ghostty, Warp, and Orca.

Claude (Anthropic OAuth/subscription) compatibility layer for pi-coding-agent — keeps ALL extension tools usable under the Claude subscription by mapping unknown flat tool names to mcp__pi__<name> on the wire instead of dropping them, plus optional multi-

为 Pi Agent 打造的计划模式 — 只读探索、计划提取与执行进度追踪 | Plan mode for Pi Agent — read-only exploration, plan extraction & execution progress tracking

Context, cache, cost, and loaded skills at a glance in Pi.

Hash-anchored read/replace/insert/grep tools for pi-coding-agent. Every line gets a unique 3-char hash (A-Za-z0-9) that stays stable across edits; stale or ambiguous anchors are rejected, never fuzzy-matched. Undo persists across restarts.

Side-call advisor primitive for Pi: consult() plus a watchdog, optional dual-path panel, and local behavior logs.

Live and final tokens-per-second meter for the pi coding agent.

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 that auto-discovers Ollama Cloud models from ollama.com and registers them as an `ollama-cloud` provider.

Confluence Cloud integration for the pi coding agent — pages, search, comments, and spaces.

SoulSpec extension for Pi Coding Agent - Load and manage AI agent personas

Pi extension adding a --stream flag: run non-interactively and stream the turn (thinking, text, and tool activity) to stdout as it happens — no TUI, no JSON.

Interactive questionnaires for Pi: tabs, multi-select, free-form typing, validation, and structured answers.

Interactive question tools for Pi — single questions, multi-step questionnaires with validation, conditional types, and open-ended input

Purpose-aware context management for long-running Prime Agent sessions

pi extension that exposes provider-native web search for DeepSeek Responses and OpenAI Codex GPT-5.6 models

Vendor-neutral Pi provider for private-ai-gateway using an instance-scoped verified ACI transport

Ask better questions. Get better answers. Rich native WebView dialogs for the Pi agent.

Execute Python code in pi as persistent background tasks

TencentDB Agent Memory integration for pi: automatic L0 capture, L1/L2/L3 recall, and memory search tools.

Only 1 tool links to Chrome — the tiniest extension you need.

LazyVim-style extension manager for Pi Coding Agent — load packages after start or on demand (cmd/tools/keywords).

Guide Pi reviewer provider diversity with same-provider fallback when alternatives are unavailable or blocked.

Pi extension that adds named subscription-account aliases for the built-in Anthropic and OpenAI Codex OAuth providers.

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

Web cockpit plugin contracts for DoomPi: plugin definitions, slot contributions, session data channels, and hub channel sources.

Non-interactive provider usage command for pi

pi extension that runs the coding-workflow (cw) dispatch loop inside a pi session

Model Context Protocol (MCP) integration for pi-coding-agent. Connect to any MCP server (stdio, SSE, StreamableHTTP, WebSocket) and use its tools, resources, and prompts natively.

Chirpi — architecture factory extension for Pi — generates project-specific architecture agents and skills from a bundled library or from your own input documents

Dev tool: on every provider request, dump the exact payload pi sends (final system prompt plus messages) to a file so you can see precisely what the model receives.

Claude-Code-style slash commands for pi: /exit and /clear.

Control your Pi coding-agent session from Discord — each session gets its own auto-created channel

Retro mainframe ambience for pi: hum + data blips while the agent thinks (70s/80s sci-fi console). One shared instance across all agents.

Pi extension that registers CodeBuddy (IOA) as an OpenAI-compatible HTTP provider with OAuth/API-key auth

Jev on OpenRouter as a tool inside Claude Code, Codex, Pi, and OpenCode: typed classify, check, score, rank, and ask, plus one-command setup.

Persistent autonomous goals for pi — /goal loops until an independent verifier confirms completion or a genuine block; audited completions, findings re-injection, auto-release

A pi coding agent extension that tracks and displays session context (worktree, Jira ticket, GitLab MR) in the footer

Pi package for structured planning, implementation, quality governance, and dashboard workflows.

Enhance Pi AI authentication, model discovery and Provider → Models selection

Zion's pi extensions: reload-all, auto-hide-thinking, codex-usage, checkpoint, commits and more