Analyze conversation for issues and automation opportunities. Use when you want to review a session for mistakes, misunderstandings, inefficiencies, or automation opportunities.
Classify a session to determine the best action (memory, rule, hook, or skill). Use when you want to quickly categorize what a session teaches us.
Create a hook/trigger from a session insight. Use when a session reveals a workflow that should happen automatically in response to events.
Create a persistent memory entry from a session insight. Use when a session reveals a user preference, project convention, or knowledge that should be remembered across sessions.
Create a coding rule from a session insight. Use when a session reveals a constraint or coding standard that should be enforced.
Analyzes a conversation transcript to identify mistakes, misunderstandings, inefficiencies, and automation opportunities. Writes analysis.json and analysis.md to the flow output directory. Use this agent when the user asks to analyze a session, review what went wrong, or find automation opportunities.
Analyzes a conversation transcript and creates a skill file to address identified issues. The agent reviews the conversation, identifies mistakes, misunderstandings, inefficiencies, or automation opportunities, and generates a SKILL.MD with trigger conditions and instructions to prevent the issue from recurring. Use this agent when the user asks to create a skill or fix a recurring problem.
Analyzes a conversation transcript and creates a Trigger entity for automated workflow actions. The agent reviews the conversation, identifies events that should trigger automatic actions, and creates a Trigger entity via the graph API. Use this agent when a session reveals a workflow that should happen automatically.
Analyzes a conversation transcript and creates a persistent memory entry. The agent reviews the conversation, identifies user preferences, project conventions, or knowledge that should be remembered, and writes it to the auto-memory directory. Use this agent when a session reveals something that should be remembered across sessions.
Analyzes a conversation transcript and creates a coding rule file. The agent reviews the conversation, identifies coding constraints or standards that should be enforced, and writes a rule markdown file. Use this agent when a session reveals a rule that should be enforced in future sessions.
Admin access level
Server config contains admin-level keywords
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A Claude Code plugin that automatically learns from your sessions. Skillit analyzes conversations to identify mistakes, inefficiencies, and automation opportunities, then creates reusable skills to prevent them from recurring.
If you don't have Claude Code installed:
npm install -g @anthropic-ai/claude-code
pip install flowpad
# macOS / Linux
curl -LsSf https://raw.githubusercontent.com/langware-labs/skillit/main/get-skillit.sh | sh
# Windows
powershell -ExecutionPolicy ByPass -c "irm https://raw.githubusercontent.com/langware-labs/skillit/main/get-skillit.ps1 | iex"
# Add the skillit marketplace
claude plugin marketplace add langware-labs/skillit
# Install the plugin
claude plugin install skillit@flowpad-ai --scope user
claude plugin list
You should see skillit in the list of installed plugins.
claude /plugin to open the plugin managerMarketplacesflowpad-ai marketplaceEnable auto-updateclaude plugin marketplace update flowpad-ai
claude plugin update skillit@flowpad-ai
If you want to develop or modify the plugin:
Clone the repository
git clone https://github.com/langware-labs/skillit.git
cd skillit
Run the installer
chmod +x install.sh
./install.sh
The installer will:
Verify installation
claude plugin list
You should see skillit in the list of installed plugins.
| Command | Description |
|---|---|
/skillit:create-skill | Launches the skill creator agent to analyze the conversation and create activation rules |
/skillit:analyze | Launches the analyzer agent to identify issues and automation opportunities |
Run /skillit:create-skill in a Claude Code session to analyze the current conversation and generate skills that prevent recurring issues:
/skillit:create-skill
Run /skillit:analyze to review a session for mistakes, misunderstandings, inefficiencies, or automation opportunities:
/skillit:analyze
Both commands launch their respective agents as background tasks, so you can continue working while they run.
npx claudepluginhub langware-labs/skillit --plugin skillitCurate auto-memory, promote learnings to CLAUDE.md and rules, extract proven patterns into reusable skills.
Core skills: ecosystem guide, skill creator, research patterns, session reflection, and plugin development. Includes UserPromptSubmit hook for forced skill evaluation.
Skill and agent authoring tools: generate, audit, and improve Claude Code skills, commands, and agents
Continuous learning hooks for gladiator MCP (observe patterns, reflect on them)
Create, test, measure, and iteratively improve Claude Code skills with category-aware design, gotchas-driven development, progressive disclosure coaching, and automated description optimization.
Automatically detect workflow patterns, generate skills, and load them dynamically mid-session