$npx skillfedfor your agent

self-learn

self-learn captures patterns, failures, and decisions from substantial sessions, scores them for generality and confidence, then proposes targeted framework updates—from project-specific steering rules to cross-cutting framework changes. It's the mechanism that lets your framework evolve from real usage, surfacing diffs for approval before any changes apply.

self-learn extracts lessons from completed sessions and proposes framework improvements based on patterns discovered.

AI-generated summary based on this skill's SKILL.md

351 54 MITupdated by Pratiyush

Decision gist · record as of 2026-06-18

self-learn extracts lessons from completed sessions and proposes framework improvements based on patterns discovered. self-learn captures patterns, failures, and decisions from substantial sessions, scores them for generality and confidence, then proposes targeted framework updates—from project-specific steering rules to cross-cutting framework changes. It's the mechanism that lets your framework evolve from real usage, surfacing diffs for approval before any changes apply.

manual: git clone https://github.com/Pratiyush/llm-wiki → cp -r llm-wiki/.claude/skills/self-learn ~/.claude/skills/self-learn
.claude/skills/self-learn/SKILL.md · version fcbaede5

Use it when

  • self-learn captures reusable patterns from sessions and converts them into steering rules that can guide future work.
  • self-learn consolidates lessons learned from your day's work by extracting patterns, failures, and decisions from substantial sessions.

Verify before relying

Read SKILL.md below before installing (1 file). Open directory: indexed for reading, not audited.

Same gist for agents: .md · .json

Install

Pratiyush/llm-wiki/self-learn · repository language: Python

Open directory. Skills are indexed for reading, not audited. Review a skill's body before installing it.

Frequently asked questions

AI-generated answers based on this skill's SKILL.md and metadata

What does self-learn do with lessons from completed sessions?

self-learn extracts lessons and patterns from completed sessions to improve your framework. It analyzes what worked, what failed, and what decisions were made, then scores those learnings for generality and confidence. The skill surfaces diffs for your approval before applying any changes, ensuring your framework evolves deliberately from real usage rather than drifting uncontrolled.

How does self-learn turn session insights into framework rules?

self-learn captures reusable patterns from sessions and converts them into steering rules that can guide future work. It distills candidate lessons from your recent debugging loops and project work, then proposes targeted framework updates—ranging from project-specific rules to cross-cutting framework changes. You review and approve each promotion before it takes effect.

What did we learn today and how does self-learn surface it?

self-learn consolidates lessons learned from your day's work by extracting patterns, failures, and decisions from substantial sessions. It scores each learning for how broadly it applies and how confident the signal is, then proposes framework improvements backed by that evidence. The skill closes the learning loop by turning session insights into actionable framework updates you can version and log.

Can self-learn promote project-specific learnings to framework level?

Yes. self-learn scores and promotes project-specific learnings to framework level when they meet your quality gates. It identifies patterns that emerged from your recent project work and evaluates whether they're general enough to become steering rules for future sessions. Approved promotions get version-bumped and logged to your changelog, creating an audit trail of how your framework evolved.

How does self-learn handle framework version updates?

self-learn version-bumps your framework after learnings are approved and applied. It logs all changes to your changelog so you have a clear record of what evolved and why. This dogfooding meta-loop ensures your framework stays documented and traceable as it grows from real session experience.

SKILL.md

Rendered from the published skill. Quoted content, verbatim.

self-learn

What this skill does

Closes the Dogfooding Meta-Loop from the Open Source Framework v4.1.

Every non-trivial session on a framework-driven project produces lessons: patterns that worked, patterns that failed, gotchas hit, decisions made. Most tools let those lessons evaporate. This skill captures them, runs a quality gate, and proposes framework updates.

It is the reason the framework evolves. Framework v4.0 → v4.1 was a self-learn pass that folded llmwiki's learnings back into the parent Open Source Framework.

When to invoke

  • User says "learn from this session", "what did we learn", "extract lessons", "distill this"
  • After a substantial feature ships (especially if it required multiple debugging loops)
  • At the

(truncated - see the full file via the links below)

File tree — 1 file
.claude/skills/self-learn/SKILL.md

Let your AI agent find skills like this

Example. Real query, live index.

You found this page by searching. An agent finds it by wishing: SkillFed indexes 56,283 agent skills by what they can do, searchable in plain language.

wish › “Extract lessons and patterns from completed sessions to improve framework”

Give your agent the search over MCP, or paste the wish link into any chat. No install? Search from any chat →

Related skills

project-maintainer
by Pratiyush · Pratiyush/llm-wiki

This skill performs a complete health check on projects following the Open Source Framework pipeline. It verifies phase gates, syncs task tracking across multiple sources, identifies stale content and broken links, and surfaces exactly what needs attention next. Use it for maintenance checks, progress updates, or whenever you need clarity on project status.

MITfor claude-codeupdated Jun 2026
★ 351repo stars
llmwiki-query
by Pratiyush · Pratiyush/llm-wiki

This skill searches your local llmwiki—a knowledge base built from your Claude Code, Cursor, and other session history—to answer questions about your own work and decisions. It reads relevant wiki pages, synthesizes answers with inline citations, and can save substantial responses as new synthesis pages for future reference.

MITfor claude-codeupdated Jun 2026
★ 351repo stars
wiki-all
by Pratiyush · Pratiyush/llm-wiki

wiki-all orchestrates the entire llmwiki workflow in sequence: initializing directories, syncing session data to markdown, constructing knowledge graphs with Graphify AI, compiling HTML output, running 14 quality checks, and launching a local server. Use it when you need to rebuild your wiki from scratch or refresh everything at once.

MITfor claude-codeupdated Jun 2026
★ 351repo stars
llmwiki-sync
by Pratiyush · Pratiyush/llm-wiki

llmwiki-sync automates the full pipeline from Claude Code session transcripts to a searchable, LLM-maintained knowledge base. It converts raw session logs to markdown, ingests them into wiki structure (sources, entities, concepts), and optionally builds a static HTML site. Use it to keep your work history organized and queryable across projects.

MITfor claude-codeupdated Jun 2026
★ 351repo stars
llmwiki-ingest
by Pratiyush · Pratiyush/llm-wiki

llmwiki-ingest transforms source files into interconnected wiki pages by extracting entities, concepts, and relationships. It writes source summaries, creates or updates entity and concept pages, cross-links everything, and flags contradictions—all while maintaining a searchable index and activity log. Perfect for building knowledge bases from session transcripts, markdown files, PDFs, and URLs.

MITfor claude-codeupdated Jun 2026
★ 351repo stars
compound-learnings
by parcadei · parcadei/Continuous-Claude-v3

Extract patterns from recent sessions and convert them into persistent capabilities—rules for heuristics, skills for workflows, and hooks for automation. The skill guides you through gathering learnings, consolidating similar patterns, detecting meta-themes, and proposing artifacts with appropriate signal thresholds before creation.

MITfor claude-codeupdated Jan 2026
★ 3,875repo stars

More skills Tailwind Ssr (unlicensed) · lab:autoresearch (MIT) · Tailwind (unlicensed)

Tags
meta-learningframework-evolutionsession-analysispattern-extractionknowledge-capturecontinuous-improvementdogfooding-loopdocumentation-automationlessons-learnedframework-versioning