Bitmap-frame context compression for vision-capable LLMs
Bitmap-frame context compression for vision-capable LLMs
Advanced compaction extension and skill for Pi with automatic threshold-based compaction and critical context preservation
Mastra-style Observational Memory extension for Pi compaction and runtime context.
Algorithmic conversation compactor for pi - transcript-preserving structured summaries, no LLM calls
CLI health checks, handoffs, visual archives, and recovery tools for OpenAI Codex session threads.
Bitmap-frame context compression for vision-capable LLMs
Improves Pi compaction by only compacting older history while keeping recent turns intact.
OpenCode plugin that automatically summarizes conversation context into structured session memory and injects it back into the system prompt every few turns.
A pi extension that caps effective model context windows at 200k tokens for earlier auto-compaction.
A pi extension that replaces default compaction with a cache-aware dynamic programming economic model
pi-di18n: full TUI i18n for pi (continues pi-i18n) plus summary localization (integrates pi-compaction-i18n)
Unified compaction + observational memory extension for Pi — compresses conversation context while preserving durable observations and reflections
Context weight-loss for Claude Code — prune bloated sessions, protect Agent Teams from compaction
Algorithmic conversation compactor for pi - transcript-preserving structured summaries, no LLM calls
CLI-first trace compaction tools for AI agent workflows.