Context-savers

Multi-phase security audits with specialist sub-agents, isolated context windows, capped concurrency, and resumable state — packaged as a Pi extension.

MCP plugin that saves 98% of your context window. Works with Claude Code, Gemini CLI, VS Code Copilot, OpenCode, and Codex CLI. Sandboxed code execution, FTS5 knowledge base, and intent-driven search.

Pi extension that keeps noisy tool output out of your context window. Automatically rewrites shell commands through Hypa for local, deterministic compression, context-aware file tools, and recoverable evidence.

Active Context Pruning (ACP) extension for Pi — model-driven context compression that keeps conversations flowing.

Headroom token compression for pi — compress large tool results via a local Headroom proxy to reclaim context window.

A pi extension that caps effective model context windows at 200k tokens for earlier auto-compaction.

Token reduction & persistent memory for Pi coding agent — compress tool outputs, save decisions across sessions, survive compaction, backed by Obsidian

Pi extension that shows context window usage visualization

Pi Coding Agent extension for OmniRoute — view combos, browse providers, and sync models with enriched metadata (context windows, max tokens, reasoning, and vision) to the Ctrl+P picker

Codex large context window extension for pi.

Rich custom status bar for pi with token usage, cost, DeepSeek balance, context window, and timing.

Progressive context compression for Pi — L0 zero-cost cleanup, L1 rule-based condensation, L2 emergency truncation, with recall mechanism.

A [pi](https://github.com/mariozechner/pi) extension for delegating tasks to subagents. Each subagent runs as an in-process SDK session with its own isolated context window — ideal for parallel work, specialist prompts, or keeping noisy subtasks out of yo

Research-backed Multi-LLM Router with parallel execution, learned routing (RouteLLM), prefix caching (RadixAttention), speculative decoding (Medusa/EAGLE), token compression (ISON), local LLM support (Ollama/vLLM/LM Studio), batch processing. Python bindi

Auto-discover local model providers (Ollama) and register their models with Pi on startup. Per-model context windows and capabilities, no /reload required.

Pi extension that automatically compacts large tool outputs before they enter the model context window, saving tokens while preserving full output as recoverable local artifacts.

Transparent LLM context compression for Pi using Headroom

Pi extension that guards context window growth by auto-limiting read and rg output

PI extension for token-efficient prose — level-based runtime output compression and markdown file compression

Pi footer/statusline token usage meter for AI coding agents, showing context window rot from smart zone to dumb zone

Token-saving proxy for the PI coding agent — intercepts and compresses API calls to reduce token consumption

Cap model context windows so pi's built-in auto-compaction fires earlier. Zero-config 200k default for long-context Claude models; configurable for anything else.

Pi extension: visual context window utilization preview (/context command)

MCP plugin that saves 98% of your context window. Works with Claude Code, Gemini CLI, VS Code Copilot, OpenCode, and Codex CLI. Sandboxed code execution, FTS5 knowledge base, and intent-driven search.

Per-model context-usage thresholds for Pi's built-in auto-compaction, so models with different context windows and performance profiles compact at the right time

Pi extension: detects and interrupts infinite thinking-block and tool-call loops in real time before they exhaust your context window.

Delegate tasks to specialized subagents with isolated context windows.

Pi coding agent extension — interactive table of your scoped models with pricing, context window, thinking levels, modalities, and coding benchmarks

Log in to the airpx LLM proxy (https://airpx.cc) with an sk-proxy API key and auto-import its model catalog (prices, context windows, reasoning, vision) into the pi coding agent.

Professional context profiler for Pi that visualizes the session context window, token distribution, and integrates with Nexus packages for actionable insights.

Filter and block models from any provider via declarative allow/block rules. Hide unwanted Copilot models, enforce global allowlists, or block by context window size.

Delegate tasks to specialized subagents with isolated context windows, structured JSON handoff, contract schemas, and live TUI streaming

Context window budget display for pi — /ctx-budget command + toggleable footer