Untested on omp
-
Independent Pi role resolution options, contributions and native startup launcher
-
Pi package for Pinet Slack assistant integration — multi-agent broker, thread routing, and inbox tools
-
Pi tool — bash shell execution with pretty output
-
Pi extension suite powered by Jev: selective context compaction and model routing, each an independently toggleable extension.
-
Async side-session questions for Pi with explicit injection into the main chat.
-
Portable Pi extension that injects Claude Code and Codex memory indexes for projects encountered during a Pi session.
-
Cognitive-driven multi-agent orchestration across Agent Kernel, Prompt Vault, ROCS, and Pi runtime owners
-
all Howaboua Pi skills
-
Pi extension: FFF-powered fuzzy file and content search.
-
Permission enforcement extension for the Pi coding agent.
-
Pi extension bundle: exit, github-tools, spec-tools, codex-usage, and opencode-usage in one install
-
Pi adapter for the optchat Rust memory engine
-
OpenCode Zen provider extension for pi
-
Shared Pi task workspace: native terminal and WebUI, persistent service, checks and human review.
-
Local web UI for the pi coding agent — one directory per workspace, in the shell of deepseek-harness
-
Portable pi agent setup: bundles the plugin collection and user config templates so another machine can be brought up with one install, with per-plugin disable/enable from the /preset command.
-
Pi skill that writes .excalidraw diagrams and opens them in a live-synced local Excalidraw window.
-
SPARK skills and runtime bootstrap for coding agents
-
Multi-agent orchestration extension for the pi coding agent: bounded worker sessions, structured episode summaries, and a risk-based review workflow.
-
pstack for Pi: rigorous agent workflows you can parallelize with confidence - poteto-mode playbooks, engineering principles, multi-model review panels, and subagents.
-
Pi coding agent extension that breaks large coding requests into tracked TODOs and runs them in bounded, adaptively reused AI worker sessions. A long-running task runner and subagent orchestrator for Pi, with a live TUI progress sidebar, retries, goal loo
-
Polished theme pack for Pi that improves contrast, mood, and visual hierarchy in the terminal UI
-
pstack for Pi: rigorous agent workflows you can parallelize with confidence - poteto-mode playbooks, engineering principles, multi-model review panels, and subagents.
-
Pi stop guard: judges premature agent stops with a second model and continues the task
-
Cross-agent installer for the feature-forge skill suite — installs the canonical forge pipeline into Claude, Codex, Copilot, Cursor, Gemini, or Pi.
-
Personal Pi extensions collected in one package, tuned to a single workflow.
-
Runs the model's bash commands as detached background tasks and wakes the agent when they finish
-
Pi TUI overlay wizard: manage models.json providers (add a relay or gateway, fetch /models, edit capabilities) and review token usage by provider/model
-
Deep vulnerability scanning of your own code, run entirely inside your Pi session at a chosen effort tier, with every finding challenged before it is reported and the verification tally computed in code. Turns surviving findings into targeted patches, eac
-
Keeps a long pi session's context small: hides old tool output, keeping each call so the agent can run it again, and writes pi's compaction summary from the session's tasks.
-
Pi metrics extension: elapsed-time HUD and resilient TPS telemetry
-
Deny-first Pi guardrails that keep LLM shell and file access safe, transparent, and user-approved.
-
A Jev (TypeSafe System One) permission gate for the Pi coding agent: it judges every bash / write / edit call before it runs, behind a config allow/deny floor and a local read-only fast path.
-
Auto mode for Pi: a classifier model, such as TypeSafe's Jev, decides which permission asks can run without a human
-
Auto-commit extension for pi-coding-agent (checkpoint-then-reorganise strategy)
-
Bridge Package for Codex and Claude Marketplace compatibility in Pi — 極簡 /codex-marketplace 純文字指令與 Headless CLI(add/list/install/update/disable/enable/remove/forget/skills)、單一 Global Bridge State、當下最新安裝與即時投影
-
Interactive questions for the pi coding agent: a single `question` tool renders option lists, tabbed forms, and free-text answers
-
Interactive clarification flows for Pi with native Herdr blocked-state support.
-
Enforce structured design→execute→finalize workflow with TDD discipline in AI coding agents
-
Allow/ask/deny guardrail extension for the Pi coding agent. No third-party runtime dependencies.
-
Active task timer extension for Pi coding agent
-
Claude Code-style dynamic workflow orchestration for Pi, built on @agwab/pi-workflow.
-
Claude as a Pi model provider through the unmodified Claude Code executable. Pi owns the transcript and tools; the CLI supplies subscription authentication as a model transport.
-
Claude auto mode for Pi & security for Pi. A zero-trust permission system for Pi coding agent powered by Advanced Secure mode to scan files, content, and tools to prevent data breaches.
-
Isolated Chromium for Pi and MCP: private approvals, saved profiles and LAN blocking
-
Pi-native tweaks for Z.AI's GLM coding models (glm-5.2, glm-5.3, glm-5.3-flash, and [1m] variants). Restricts the Pi thinking-level UI to the modes each model supports on the wire, wires the native thinkingFormat:"zai" translation, and auto-clamps hidden
-
Ask-user-question tool for pi: lets the agent present multiple-choice questions to the user.
-
pi package exposing the @rprtr258/md CLI as agent tools
-
Web search, deep research and fetch content tools for Pi Agent, powered by Exa.
-
A pi extension that dynamically prunes stale context to keep sessions lean.