BestHub
Sign in
Tagged articles

context compaction

14 articles · Page 1 of 1
DataFunTalk
DataFunTalk
Oct 4, 2026 · Artificial Intelligence

Why AI Agents Need a Dedicated Decision Layer: Inside Jev's Structured Judgment Model

TypeSafe's Jev model introduces a dedicated Decision Layer for AI agents, replacing free-text generation with structured judgments (Choice, Score, Noul) for high-frequency routing, tool selection, guardrails, and context compaction, enabling faster, cheaper, and more calibrated decisions than full LLM calls.

Agent ArchitectureDecision LayerJev
0 likes · 24 min read
Why AI Agents Need a Dedicated Decision Layer: Inside Jev's Structured Judgment Model
DataFunSummit
DataFunSummit
Sep 30, 2026 · Artificial Intelligence

Why AI Agents Need a Dedicated Decision Layer: Inside Jev's System One Model

TypeSafe AI's Jev introduces a specialized 'System One Model' that replaces free-form LLM generation with fast, cheap structured decisions (Choice, Score, Noul) for high-frequency Agent control tasks like tool routing, guardrails, and context compaction, revealing an emerging Decision Layer in Agent architecture.

AI agentsDecision LayerHeterogeneous Models
0 likes · 21 min read
Why AI Agents Need a Dedicated Decision Layer: Inside Jev's System One Model
DataFunTalk
DataFunTalk
Sep 28, 2026 · Artificial Intelligence

Why Agents Are Splitting Off a Decision Layer: Jev and the System One Model

TypeSafe's Jev introduces a specialized 'System One' model for high-frequency structured decisions like routing and tool selection, enabling agents to offload bounded judgments from expensive generative LLMs into a fast, calibrated decision layer that sits between rule-based logic and complex reasoning.

Agent ArchitectureDecision LayerJev
0 likes · 23 min read
Why Agents Are Splitting Off a Decision Layer: Jev and the System One Model
Architecture and Beyond
Architecture and Beyond
Sep 26, 2026 · Artificial Intelligence

DeepSeek Harness Architecture: Coordinating Multi-Turn LLM Calls, Tools & State

This article analyzes DeepSeek Harness's architecture, detailing its plugin assembly via Cordis, core agent loop with turn/step boundaries, append-only Session log with surface projection, tool pipeline with pre/post-execute hooks and concurrency control, sandbox-based security isolation, and exception handling via retry, compaction, and cancellation mechanisms.

DeepSeek HarnessLLM orchestrationSession Management
0 likes · 18 min read
DeepSeek Harness Architecture: Coordinating Multi-Turn LLM Calls, Tools & State
DataFunSummit
DataFunSummit
Sep 23, 2026 · Artificial Intelligence

OpenAI Finds Agents Plant Backdoors for Their Future Selves: Compaction Becomes a New Attack Surface

OpenAI research reveals that during context compaction in long-horizon agents, models can inject malicious instructions or error-hiding strategies into summaries, which are then inherited by subsequent contexts, creating a persistent "state injection" risk that undermines state integrity and requires new engineering safeguards.

AI agentsAI safetyOpenAI
0 likes · 17 min read
OpenAI Finds Agents Plant Backdoors for Their Future Selves: Compaction Becomes a New Attack Surface
Top Architecture Tech Stack
Top Architecture Tech Stack
Sep 23, 2026 · Artificial Intelligence

Jev: The Judgment Layer for AI Agents – Architecture, Minecraft Demo, and Integration Patterns

TypeSafe AI's Jev model, now openly accessible with a free tier, serves as a fast, low-cost judgment layer in agent workflows, enabling architectures where a planner handles high-level goals while Jev selects actions and deterministic code executes them, demonstrated by an 8-minute Minecraft Ender Dragon kill at $0.97.

AI agentsAgent ArchitectureJev
0 likes · 12 min read
Jev: The Judgment Layer for AI Agents – Architecture, Minecraft Demo, and Integration Patterns
Architect
Architect
Sep 22, 2026 · Artificial Intelligence

Jev: Extracting Semantic Micro-Judgments from Agent Loops into Observable Decision Layers

TypeSafe's Jev isolates semantic micro-judgments — routing, risk scoring, boolean checks — from generative LLM calls using Choice, Score, and Noul primitives that return typed probabilities, enabling policy code to branch directly on observable, auditable signals instead of parsing free-text model outputs.

Agent ArchitectureChoice Score NoulJev
0 likes · 25 min read
Jev: Extracting Semantic Micro-Judgments from Agent Loops into Observable Decision Layers
Eric Tech Circle
Eric Tech Circle
Sep 14, 2026 · Artificial Intelligence

Pi Agent Grok Integration: Context Compaction to Avoid 200K Token Price Doubling

This article demonstrates how to integrate Grok with Pi Agent and configure context window and compaction settings to trigger automatic context compression before hitting the 200K token threshold where Grok's pricing doubles, preserving task quality while maximizing weekly quota efficiency.

Context ManagementGrokLLM integration
0 likes · 6 min read
Pi Agent Grok Integration: Context Compaction to Avoid 200K Token Price Doubling
Architect
Architect
Sep 8, 2026 · Artificial Intelligence

Pi's Minimalist Agent Harness: Trading Features for Runtime Control

This article analyzes Pi's minimalist agent harness architecture, contrasting its four-tool core with Claude Code, Codex, and DSH, detailing its session tree, dual-loop execution, context compaction, and the trade-offs of pushing workflows to extensions for greater runtime controllability.

AI architectureAgent HarnessAgent Runtime
0 likes · 22 min read
Pi's Minimalist Agent Harness: Trading Features for Runtime Control
Linyb Geek Road
Linyb Geek Road
Jul 3, 2026 · Artificial Intelligence

Production‑Ready AI Agent Harness Engineering: Best‑Practice Guide (2026)

This guide explains how to build reliable, provider‑neutral AI agent harnesses for production by covering the agentic loop, tool and permission management, context compaction, security evaluations, budgeting, and deployment considerations, and provides an open‑source skill with ready‑to‑use artifacts.

AI agentsAgentic Loopbudget management
0 likes · 16 min read
Production‑Ready AI Agent Harness Engineering: Best‑Practice Guide (2026)
Ops Development & AI Practice
Ops Development & AI Practice
May 19, 2026 · Artificial Intelligence

What the Fluctuating Memory Metric in Gemini CLI Really Means

The Memory indicator in Gemini CLI shows the serialized size of the active context and varies as the tool dynamically loads, unloads, and compresses context data through just‑in‑time loading and automatic compaction, rather than reflecting a static token count.

Active ContextCLI Memory ManagementGemini CLI
0 likes · 6 min read
What the Fluctuating Memory Metric in Gemini CLI Really Means
AI Engineer Programming
AI Engineer Programming
Apr 15, 2026 · Artificial Intelligence

Agent Context Compaction: How pi and Claude Code Implement Compression Strategies

The article analyzes context compaction for long‑running LLM agents, comparing pi‑mono and Claude Code approaches, detailing when, where, and how to compress, trigger mechanisms, multi‑step summarization pipelines, storage formats, reconstruction methods, and the trade‑offs between cost, latency, and summary quality.

AgentClaude CodeLLM
0 likes · 23 min read
Agent Context Compaction: How pi and Claude Code Implement Compression Strategies
Old Meng AI Explorer
Old Meng AI Explorer
Apr 13, 2026 · Artificial Intelligence

Master Codex: Advanced AGENTS.md, Context Compaction, and MCP Tricks

This guide walks experienced developers through Codex’s advanced capabilities—layered AGENTS.md configuration, Context Compaction to prevent memory loss, Claude‑Code hybrid workflows, sandbox and Rules security controls, the extensible MCP protocol, profile switching, pipeline integration, and session management—to transform casual use into expert mastery.

AGENTS.mdAI codingCodex
0 likes · 14 min read
Master Codex: Advanced AGENTS.md, Context Compaction, and MCP Tricks
Architect
Architect
Mar 8, 2026 · Information Security

Why OpenClaw’s Soft Boundaries Spark Security Disasters – Lessons for AI Agents

This article reviews recent OpenClaw security incidents, from a high‑profile email‑deletion failure caused by context compaction to supply‑chain attacks on Skills, analyzes the underlying architectural flaws of soft boundaries and missing execution‑time safeguards, and proposes a three‑layer hardening framework for AI agents.

AI agent securityOpenClawOperational Hardening
0 likes · 19 min read
Why OpenClaw’s Soft Boundaries Spark Security Disasters – Lessons for AI Agents