Store and recall ADRs, phase context, and learnings across SDLC sessions.
npx skills add https://github.com/arbgjr/smart_alarm --skill memory-manager-arbgjr
Please help me install this Agent Skill. Skill: memory-manager Source: https://github.com/arbgjr/smart_alarm/tree/main/.claude/skills/memory-manager Command: npx skills add https://github.com/arbgjr/smart_alarm --skill memory-manager-arbgjr
Memory-manager provides persistent project context, ADRs, and learnings to maintain continuity across sessions.
Create a new project memory space and begin storing ADRs, phase context, and learnings. Then remember to migrate from legacy memory on first use.
High-intent search queries and answers about installing and using this skill.
To maintain project context across sessions, you can use a persistent storage utility to save phase context, ADRs, and learnings. This ensures continuity by storing artifacts in a structured directory layout for later recall.
Tracking architectural decisions and learnings is best handled by saving them as ADRs and indexed learnings within a persistent project memory structure. This allows you to recall specific phase context and historical decisions during later development stages.
Yes, you need Python and the PyYAML dependency to operate these project memory and ADR management scripts. The utilities rely on PyYAML to parse and store artifacts within the structured memory layout.
Yes, you can migrate legacy project memory into the new structure. The memory management utilities support migrating from legacy memory layouts on first use, ensuring your historical ADRs and phase context are preserved in the new directory format.
Saving ADRs and phase context improves development continuity by persisting decisions and environmental states across sessions. Indexed learnings and recalled context prevent knowledge loss, allowing future sessions to resume work with full historical awareness.
File-based decision tracking relies on a local directory structure, meaning artifacts are stored as files requiring PyYAML to parse. Without a centralized database, managing cross-project memory requires manual synchronization of these local storage directories.
yaojingangAutomate creation, evaluation, packaging, and governance of reusable agent skills.
nnexaiRetrieve historical project decisions and patterns from the MemPalace knowledge base.
HHU3637krStore and index experience and knowledge memories into spec context directories.
jwilgerStore and retrieve agent knowledge in markdown files with grep-based search.
