AI-powered website builder that generates a professional site in under a minute and includes tools for content, SEO, and analytics.
52M visits/mo+0.7%#4 Coding & Development
threadle
Local-first tool that turns AI coding sessions into inspectable, reusable workflows
22Popularity
About threadle
threadle converts local AI coding sessions into visual graphs that can be inspected, edited, and rerun. It captures transcripts, tool calls, token usage, and file changes from agents such as Claude Code, Cursor, or Copilot. Sessions are grouped by project with live status indicators. Each session is rendered as a graph showing turns, tools used, and files touched, with subagents branching out. Users can create reusable workflows by wiring prompts, agents, and context into a canvas that executes on the server. The tool tracks lineage of briefs across sessions and tools, allowing context to be reused or injected into new providers. Runs are listed with status, duration, and results, and can be canceled or replayed. Full-text search across transcripts and payloads is supported with ranked snippets.
Key features
- Session-to-graph visualization with turns, tools, and files
- Canvas for wiring prompts, agents, and context into reusable workflows
- Lineage tracking for briefs across sessions and tools
- Timeline view with provider-colored swimlanes
- Full-text search across transcripts and payloads
- MCP server integration for in-session queries
- Run management with cancel, replay, and status tracking
- Project-level grouping with live status indicators
Use cases
- Debugging and reviewing AI coding sessions with visual context
- Reusing workflows across different AI agents and projects
- Tracking token usage and spending per provider and project
Pros
- Local-first architecture with read-only access to agent storage
- Visualizes sessions as interactive graphs with lineage tracking
- Supports multiple AI coding agents and MCP integration
- Enables reusable workflows and parallel task execution
- Provides full-text search and statistics on token usage and spending
Cons
- Requires local installation and runs on a specific port
- No cloud-based version or hosted service available
- Limited to agents that write session data locally
Frequently asked questions about threadle
What is threadle and what does it do?
threadle converts local AI coding sessions into visual graphs that can be inspected, edited, and rerun. It captures transcripts, tool calls, token usage, and file changes from agents such as Claude Code, Cursor, or Copilot, then renders them as interactive graphs.
Who is threadle designed for?
threadle is designed for developers and teams who use AI coding agents and want to track, analyze, and reuse their work. It suits those who need visibility into AI session details, reusable workflows, and cross-tool context sharing.
How does threadle integrate with existing AI coding tools?
threadle integrates with local AI coding agents like Claude Code, Cursor, and Copilot by reading their session transcripts and tool calls. It also exposes saved workflows as tools via the MCP protocol, allowing agents to query threadle directly.
Can I reuse workflows or contexts across different AI providers?
Yes, threadle allows users to distill sessions into reusable briefs and inject them into different providers while preserving context lineage. Workflows can be wired into a canvas and executed on the server.
Is threadle a cloud service or does it run locally?
threadle runs locally and is not a cloud service. It operates in read-only mode against agent storage and does not mutate local files. The tool is installed locally and accessible via a local web interface.
How do I get started with threadle?
Getting started with threadle involves installing it locally using a provided command or cloning the repository. Once installed, sessions automatically appear, and users can begin exploring graphs, workflows, and runs through the local web interface.
