prompttool
pi-cache-optimizer
Improve Pi prompt/KV cache hit rates with stable prompts, OpenAI-compatible cache keys, proxy compat warnings, and footer cache stats.
Packages
6.2k listings for pi · 0 reviewed by us
◒ Models & providers Local models, alternative providers, routing, usage and cost meters.
29 results
prompttool
tokenomy-pi
Automatic token-saving router for the Pi CLI using Pi's openai-codex provider, with visible local usage and quality evidence.
+4
promptextension
pi-dialer
Automatic per-prompt model and thinking-level routing for pi
+1
prompttool
@ersintarhan/pi-toolkit
pi extension that registers Kimi, MiniMax, Xiaomi MiMo, and CrofAI providers with cache fix, Anthropic OAuth adapter, and native web search with provider-override support. Search backends include ZAI MCP, Google, OpenAI, xAI, Anthropic, Claude Code, and C
+5
promptextension
pi-star-router
StarRouter is a focused model-routing extension for Pi. It chooses the best available model and thinking level for each prompt using a deterministic heuristic prompt profiler, Artificial Analysis benchmark data, provider-scoped model pools, strict model i
prompttool
pi-opencode-go-cache
Prompt caching for Pi's OpenCode Go provider (kimi, deepseek, mimo, qwen, minimax). Stamps prompt_cache_key, 24h retention, and cache_control breakpoints on every request — beyond what pi-ai or opencode CLI do by default.
promptextension
pi-gemini-acp
Gemini ACP chat, prompt, search, and research provider for Pi.
+3
promptextension
pi-for-k3
Kimi K3 prefix-cache optimizer extension for pi-agent (dual-channel, zero-fork)
+4
promptextension
pi-streamlake-provider
StreamLake AI custom provider extension for pi. Registers two provider entries (streamlake-coding / streamlake-payg) with baked-in base URLs, no redundant billing prompt, and dynamic model list fetch from the API.
prompt
pi-llama-cpp-stats
Shows prompt processing progress from llama.cpp's SSE stream
promptextension
@guygrigsby/pi-promptdump
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.
promptextension
pi-prompt-composer
Build multi-option slash commands from plain prompts — variable expansion, arg collection & interactive selectors for Pi
prompttool
pi-tscg
Drop-in tool-schema and tool-result compression for the Pi coding-agent. Built on TSCG by Furkan Sakizli (SKZL-AI) — wraps @tscg/core and adds tool-result compression plus provider-aware prompt-cache awareness (Anthropic/OpenAI/Ollama). Provider-agnostic,
promptextension
pi-screenshots-picker
Screenshot picker extension for pi coding agent - quickly select and attach screenshots to your prompts
promptextension
pi-cost-switch
Pi extension: preview next-turn model/thinking cost with hit, cold, and cache-tax estimates before switching
promptextension
@mcowger/pi-better-messages-cache
Pi extension: dual cache-breakpoint strategy for Anthropic models — marks both the last assistant tool_use block and the last user message block with cache_control, dramatically improving cache hit rates on MiniMax, Kimi, and other Anthropic-compatible pr
promptextension
@pmlpm/pi-translate-zh-en
Pi extension: bidirectional zh<->en translation to save tokens. Translates user Chinese input to English before sending to the model, and translates the model's English response back to Chinese. All-in-one replacement for pi-prompt-translate (input side)
+2
prompt
pi-openai-instructions-provider
Pi provider package for OpenAI-compatible Responses APIs that sends system prompts via instructions and uses a stable prompt_cache_key.
promptextension
@samfp/pi-steering-hooks
Deterministic tool-call guardrails for pi — enforce rules with before-tool hooks instead of prompts. Zero token cost, 100% reliability.
promptextension
@furkanbilgin/pi-retry
Auto-retry Pi extension — automatically recovers from transient provider errors (400, 5xx, rate limits, connection failures) mid-conversation by sending a continuation prompt with exponential backoff
prompt
@gnoviawan/pi-session-usage
Tracks time usage, prompts, turns, and tool calls in a session.
promptextension
@artale/pi-agent
Agentic engineering toolkit. Token estimation, cost, patterns, templates.
prompt
pi-prompt-tracker
Track complete prompts sent to the LLM gateway — system prompt, tools, messages, responses, and usage
promptextension
pi-model-provider-native-prompting
Provider-informed, harness-neutral native prompting guidance for Pi Coding Agent.
promptextension
pi-minimax-usage
Minimax M3 token usage, cost, prompt-cache savings, and 5h/weekly subscription quota in the Pi status line
promptextension
pi-gpt-cache
Pi extension that adds prompt_cache_key to OpenAI Responses API requests for better prompt caching.
promptextension
pi-auto-mode-router
Auto Mode Router extension for Pi: route prompts between frontend, logic, and terminal models with optional mid-turn domain switching.
+3
promptextension
pi-cache-ttl-config
Dynamically switch the Anthropic prompt-cache TTL between 5m and 1h mid-conversation in the pi coding agent
promptextension
pi-session-router
Pi extension that uses an LLM to route new prompts into the most relevant existing session before continuing
+2