Tom

Terminal multiplexer with master-stack tiling layout

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

Anthropic Agent Skills Specification plugin for OpenCode - discover and execute skills with dynamic tool registration

Improve OpenAI in pi with fast mode, usage stats, realtime voice, image generation, and footer polish.

Replay LLM plugin: short-circuits llm/stream with model chunks reconstructed from a recorded session JSONL (keyless snapshot tests)

localterm <-> pi integration: Kitty graphics for the browser renderer, scrubbing localterm-managed secret env vars from the agent's bash-tool children, and OSC 9 desktop notifications when the agent finishes a long turn.

[](https://codecov.io/gh/monotykamary/flame-js)

A browser-based terminal: one browser tab is one PTY session. Friendly URL, persistent xterm.js front-end, hono + node-pty back-end.

Validate-then-repair extension for pi — fixes common LLM tool-call mistakes (null fields, stringified arrays, wrong field names, anchor bleed) before tools execute

Neuralwatt provider extension for pi - Access Kimi, GLM, Qwen, GPT-OSS, Devstral, and MiniMax models through the Neuralwatt API

Sort models in pi's /model selector by last usage — most recently used models appear first

Namespace tools and skills in pi — group tools by extension with configurable prefix rewriting

Makora provider extension for pi - Access DeepSeek V4, GLM 5.1, Kimi K2.6, Llama 3.3, Qwen 3.6, and more through the Makora inference API

Autonomous experiment loop for pi — run, measure, keep or discard. Inspired by karpathy/autoresearch.

Skip the startup trust prompt in pi — auto-decline project trust so you're immediately interactive, then /trust + /reload when ready

Give text-only pi models vision — describe images with a vision model you pick via an interactive picker, then hand off the text description to non-vision models

Hide providers and models from pi's model selector — filter the /model list and Ctrl+P cycling via a configurable blocklist

Tokens-per-second tracker for pi — see your LLM generation speed after every agent turn

Algorithmic conversation compactor for pi - transcript-preserving structured summaries, no LLM calls

Baseten provider extension for pi - Access DeepSeek, Kimi, GLM, MiniMax, Nemotron, and GPT-OSS models through the Baseten Model API

Fireworks AI provider extension for pi - Access Kimi, MiniMax, GLM, DeepSeek, and GPT-OSS models through the Fireworks AI API

IO Intelligence provider extension for pi - Access DeepSeek, Kimi, GLM, Llama, Qwen, Mistral, and more through the IO Intelligence API

Tensorix provider extension for pi - Access DeepSeek, GLM, Kimi, Qwen, MiniMax, Llama, and more through the Tensorix API

CrofAI provider extension for pi - Access Kimi, GLM, Qwen, DeepSeek, MiniMax, and Gemma models through the CrofAI API

HyperCharm provider extension for pi - Access DeepSeek, GLM, Kimi, Qwen, MiniMax, Gemma, and GPT-OSS models through the Charm Hyper API

Parasail provider extension for pi - Access top open-source models (DeepSeek, Qwen, GLM, Kimi, Llama, Gemma, Mistral) through the Parasail API

macOS computer-use for Pi via harness server and CLI.

DeepSeek provider extension for pi - V4 Pro, V4 Flash & Reasoner with prefix cache optimization

Opencode Go provider extension for pi - Fast, efficient GLM, Kimi, and MiniMax models through the opencode.ai API

Lilac provider extension for pi - Access Kimi K2.6, GLM 5.1, and Gemma 4 models through Lilac's OpenAI-compatible API on idle GPUs

Billing engine for the serverless agency — a pi extension that meters agentic dev work like cloud compute and invoices it like a timesheet.

Zyloo provider extension for pi - Claude, GPT, Gemini, Grok, Kimi, GLM, DeepSeek, Qwen & MiniMax through the Zyloo OpenAI-compatible API

Merius provider extension for pi - Open models on Merius's own GPUs via an OpenAI-compatible API

UMANS provider extension for pi - Access Kimi K2.6, GLM 5.1, Qwen 3.6, and coding-optimized models through the UMANS API

Token-budgeted repo mapping for agent sessions: foveated heat diffusion over a cross-language code graph, with progressive disclosure.

Moonshot AI provider extension for pi - Access Kimi K2.6, K2.5, K2 Thinking, and Moonshot V1 models through the Moonshot API

Recursive agent extension for Pi — spawn subagents programmatically with depth guardrails

Lazy-load pi extensions on demand via a ToolSearch-style proxy tool

Opencode provider extension for pi - Access GPT, Claude, Gemini, GLM, MiniMax, Kimi and more through the opencode.ai API

Provider-neutral LLM service interface for the DeepSeek Harness

Storage hub (ctx.storage): named backend registry plus mounted data-form facilities for the DeepSeek Harness

Domain data form (ctx.storage.domain): schema-validated, event-emitting KV domains over storage backends for the DeepSeek Harness

DeepSeek chat-completions adapter for the DeepSeek Harness LLM seam

pi-ai-backed DeepSeek adapter for the DeepSeek Harness LLM seam (design-verification twin of dsh-llm-deepseek)

Provider-routed LLM request retry policy for the DeepSeek Harness

Shared LLM generation policy for DeepSeek Harness session-title providers

First-message LLM provider plugin for DeepSeek Harness session titles

JSON file KV storage backend for the DeepSeek Harness storage hub

Scriptable OpenAI-compatible HTTP/SSE fault server for LLM recovery tests

All-user-messages LLM provider plugin for DeepSeek Harness session titles

SQLite storage backend (kv facet) for the DeepSeek Harness storage hub

Foveated repository intelligence for DeepSeek Harness

Fabric-owned system prompt override and native-prose minimization for DeepSeek Harness

Leader-elected Factory scheduler and DSH Agent executor

Zro provider extension for pi - GLM-5.2, GLM-5.3 Flash, Kimi K3 & DeepSeek V4 Flash through the Zro inference endpoint

Schema-guided tool-call repair for DeepSeek Harness

Describe images with a vision model so text-only DeepSeek Harness models can read them

Invisible session continuation for pi — resume the agentic loop without sending ANY prompt the LLM can see

CoralBricks provider extension for pi - Access GLM 5.2/5.3, Kimi K3, and GPT-OSS 120B through the Coral Inference API

Render LaTeX as terminal images in Pi's TUI.

Same-provider multi-account pooling, OAuth storage, and safe in-stream auth failover for Pi

ChatGPT OAuth, Codex models, search, read_image URL support, and gpt-image-2 generation for DeepSeek Harness

Toggle skill visibility in pi's system prompt — flip disable-model-invocation on skills via an interactive TUI

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

Pareto Inference provider extension for pi - GLM-5.3 Flash, GLM-5.3, and DeepSeek V4 Flash through Pareto's OpenAI-compatible API

DeepSeek-backed search provider (native web_search via the Anthropic-compatible API) for the DeepSeek Harness web capability seam (ctx.web)

Trajectory event ledger with an interactive timing overview: pure-consumer plugin registering into the conversation ViewMap (no service)

Abstract spill storage seam (ctx.spillStore) for the DeepSeek Harness — save oversized tool text and return a retrieval locator

Perplexity-backed search provider for the DeepSeek Harness web capability seam (ctx.web)

Abstract web access capability seam (ctx.web) for the DeepSeek Harness — search/fetch provider registry, registration-order-independent selection, request/result vocabulary, and the WebError taxonomy