Captures session lessons into Azoth's memory system and classifies episodes for promotion.
npx skills add https://github.com/yiwei79/azoth --skill remember-yiwei79
Please help me install this Agent Skill. Skill: remember Source: https://github.com/yiwei79/azoth/tree/main/.opencode/skills/remember Command: npx skills add https://github.com/yiwei79/azoth --skill remember-yiwei79
Capture durable lessons from experience. Every session produces knowledge — this skill ensures it's not lost.
Run the remember capture at session end to archive a new episode and start classification
High-intent search queries and answers about installing and using this skill.
Capturing durable lessons from a session is automated by appending memory episodes to an append-only log. This ensures session knowledge is retained durably without manual effort, classifying lessons for future reuse.
Classifying session lessons involves auto-detecting patterns and applying governance rules for M3→M2 memory tier transitions. This promotion pathway surfaces reusable knowledge while enforcing governance over your memory system.
Memory episode capture should be triggered at session end or after major decisions. Running the capture at these points archives the episode and starts classification, ensuring critical context is preserved.
Historical memory episodes cannot be mutated because the system uses an append-only log. You can apply episodic remediation and add new linked entries, but original historical data remains unchanged to preserve auditability.
The memory system supports tagging and linking for session episodes. This governance-friendly surfacing connects related episodes to support future recall and provide decision justification across sessions.
yiwei79Captures session lessons as structured episodes with append-only governance and cross-linkages.
erikliebenExtract reusable patterns and rules from session history and decision logs.
MesferAliRun LLM inference on CPUs, Apple Silicon, and non-NVIDIA GPUs with GGUF quantization.
Kushal9889Extract and aggregate writing fragments into a single Markdown file.
