Manage persistent agent context through structured focus sessions and daily logs.
npx skills add https://github.com/F0WCUS/memory-focus --skill memory-flow
Please help me install this Agent Skill. Skill: memory-flow Source: https://github.com/F0WCUS/memory-focus/tree/main/skills/memory-flow Command: npx skills add https://github.com/F0WCUS/memory-focus --skill memory-flow
This skill solves the productivity-killing issue of context loss between agent sessions by providing a structured framework for tracking work, decisions, and long-term memory.
Initiate your work session by typing BEGIN to start the setup wizard and load your project context.
High-intent search queries and answers about installing and using this skill.
Persistent agent context is maintained by logging decisions and open loops into a centralized repository. This requires filesystem access to synchronize daily logs and state markers for session continuity.
The best way to track decisions and open loops is through structured FOCUS sessions. This workflow provides BEGIN and END commands to tag active tasks and manage long-term memory retention automatically.
Start a focus session by typing BEGIN to trigger the setup wizard. This action loads your project context and initializes the daily log synchronization needed for the current work block.
Yes, filesystem access is required to manage agent memory logs. The system needs to maintain state markers, daily logs, and configuration files directly in the repository to ensure session continuity.
Continuous improvement in memory tracking is driven by a 1% Better Loop. The system prompts reflection every five sessions to encourage ongoing optimization of your automated workflow and context retention.
Agents lose project status because they lack a structured framework for context retention. Without cycling context through daily logs and a central memory file, previous decisions are not retained across interactions.
yocxy2Restore session context and workspace state for multi-session engineering tasks.
bibo1243Extract and codify daily work decisions into a centralized organizational knowledge base.
dotBeepsDisplay tag-grouped todos as persistent floating panels in Pi agent workflows.
TbK0ngRecord work sessions with context and memory updates in Trellis projects.
