context-compaction

1207 public resources for Jev, TypeSafe AI's System One decision model, indexed by decision pattern. Source citations, dated link checks and scheduled call-site text checks; runtime and performance are not independently tested here. EN/中文, JSON schema and platform compatibility.

Pre-compaction hooks + a CAS-guarded task pointer that keep Codex agents on-goal across context auto-compaction — no mainline drift, no duplicate work, no conflicting progress in long runs.

Zero Code Studio 是一个工程化 AI 代码生成平台，基于 Spring Boot、LangChain4j、LangGraph4j，支持工具调用、Workflow 编排、任务计划规划、三层上下文压缩、SSE 过程可视化、定点修改与一键本地部署。

TypeScript CLI: extract any coding-agent session (Claude Code, Codex, Pi) into a compact, verified, provenance-linked handoff another agent can continue from.

A local GUI workaround for Codex Desktop conversations stuck on automatic context compaction.

Six in-process middlewares for OpenClaw: HITL approvals, prompt-injection guardrails, PII redaction, tool-call budgets, context compaction, and complexity-aware model routing. Zero telemetry, all state local.

Aggressive context compaction for local-first agents. Runs Qwen3.8‑27B on self‑hosted llama.cpp at low context, shrinking history so the live prompt stays small, fast, and private—delivering a big‑window experience without API cost or data egress. 面向本地的激进上下文压缩插件。自托管 llama.cpp 低上下文运行 Qwen3.8‑27B,不断收缩历史、保持常驻 prompt 小而快,兼顾隐私与大窗口体验,零 API 成本、数据不出本机。

Cache-friendly context compaction for OpenCode: summarize by append, then cut the outgoing context.

Gobstopper makes long Claude Code and Codex sessions smaller. A local proxy compacts live requests, and saved sessions get a smaller copy beside the original.

GenPark AI Agent Skill - Streaming attention sink preserver and FIFO sliding-window KV cache eviction manager for infinite-context agent loops.

GenPark AI Agent Skill - Streaming attention sink preserver and FIFO sliding-window KV cache eviction manager for infinite-context agent loops.

Pauseless compaction for Claude Code: long sessions summarised in the background with no pause. Local gateway: your Claude subscription first, overflow to Together AI (GLM), OpenRouter, any OpenAI-compatible endpoint, Bedrock or an API key. Keeps Remote Control working.

An experimental terminal coding agent with configurable models, Witch orchestration, and multiple context compaction methods.

Multi-persona deliberation and Bayesian weighted consensus engine resolving complex agent decision ambiguities

Long-horizon context window compression and semantic deduplicator eliminating redundant tool outputs

Multi-turn autonomous agent trajectory evaluator measuring step redundancy and tool choice efficiency

Pre-commit diff safety scanner preventing destructive shell commands, API token leaks, and disk corruptions

Long-horizon context window compression and semantic deduplicator eliminating redundant tool outputs

Multi-persona deliberation and Bayesian weighted consensus engine resolving complex agent decision ambiguities

Fault-tolerant tool execution circuit breaker detecting cascading upstream errors and triggering fallback routing