@DanKornas: Your coding agent shouldn’t start from zero every session Magic Context is a CortexKit plugin for giving coding agents …
Summary
Magic Context is an open-source CortexKit plugin that gives coding agents self-managing context and long-term project memory across OpenCode and Pi, allowing persistent sessions and automatic memory capture.
View Cached Full Text
Cached at: 06/20/26, 04:20 PM
Your coding agent shouldn’t start from zero every session
Magic Context is a CortexKit plugin for giving coding agents self-managing context and long-term project memory across OpenCode and Pi.
It helps you keep one long-running project session alive by moving older history into tiered compartments, capturing durable project knowledge, and recalling the right memories when the agent needs them.
Key features:
• Background context management – historian compresses older session history into scored compartments so the active window stays small • Automatic memory capture – lifts decisions, constraints, conventions, config values, and naming rules into project memory • Optional overnight consolidation – dreamer merges duplicates, verifies memories against the codebase, archives stale facts, and maintains docs • Cross-session recall – searches project memories, raw conversation history, and opt-in indexed git commits when you need old context • OpenCode + Pi support – setup wizard detects harnesses, configures plugins, and shares one CortexKit database across both
It’s open-source (MIT license).
Link in the reply
Similar Articles
@DanKornas: Agents shouldn’t make the same mistake every session. Agentic Context Engine (ACE) is an open-source Python engine for …
Agentic Context Engine (ACE) is an open-source Python tool that adds persistent learning to AI agents via a Skillbook of strategies refined from execution traces and feedback.
What I'm learning trying to ensure context continuity for different agents across different sessions
The author introduces AICTX, an open-source tool that preserves structured operational state across coding agent sessions, reducing the need for agents to rediscover repository context each time.
Agent Context
Agent Context is a dev tool that lets users attach reference projects to AI coding assistants.
Maybe Coding Agents Don't Need a Bigger Memory. Maybe They Need Continuity
The article argues that coding agents need continuity—preserving execution history and project state in the repository—rather than simply larger memory or context windows, to avoid losing the operational thread between sessions.
I built an open-source coding agent that makes context visible and editable — you curate exactly what the LLM sees
The author built Nice Coding Agent, an open-source coding workbench with a visible and editable context stack, allowing users to curate exactly what the LLM sees. It features local-first retrieval, sandboxed execution, and hybrid code search, aiming to give developers control and visibility over context assembly.