Capture session learnings into memory, wiki, and team repositories.
npx skills add https://github.com/agentteamland/core --skill save-learnings
Please help me install this Agent Skill. Skill: save-learnings Source: https://github.com/agentteamland/core/tree/main/skills/save-learnings Command: npx skills add https://github.com/agentteamland/core --skill save-learnings
Save learnings at the end of conversations across multiple persistence layers: the project memory, the mandatory wiki, the team repository, and the journal. It also auto-creates new skills, agent children files, and rules when repeating patterns are detected, removing the need for confirmation and enabling autonomous improvements.
Run /save-learnings at the end of a session or with --from-markers to automatically collect learnings into memory, wiki, and the team repo.
High-intent search queries and answers about installing and using this skill.
You can automatically save agent learnings to a team wiki and repository by running a session-end capture that extracts knowledge, updates project memory, pushes changes to the repo, and generates an activity report.
The best way to capture conversation learnings without manual confirmation is using an autonomous process that detects repeating patterns and auto-creates new skills, rules, and agent children files directly from the session transcript.
Marker-driven extraction works by parsing specific comment tags in your transcript to focus processing, reducing overhead by isolating only the marked sections to summarize and distribute across memory and wiki layers.
Yes, you can auto-create new skills and agent rules from repeating patterns by enabling autonomous improvements that analyze session data and generate children files without requiring manual confirmation.
Yes, end-of-session knowledge capture enforces mandatory wiki updates to maintain current knowledge across the project, alongside optional journal logging and doc-sync to ensure distributed truth.
You should not use marker-based extraction when your transcript lacks specific comment tags, as the parser relies on these markers to scope processing and reduce overhead during the learning capture workflow.
Ar9avIngest Claude conversation histories into Obsidian wiki pages organized by project context.
pddhktCapture session learnings and prepare commit-ready updates for the kotlin-kmp-skills repository.
ZESCODESynchronize project knowledge across local files, MCP memory, and external repositories.
scooter-lacroixExtract and categorize patterns from session learning cache files.
