
Threadmap Mcp
io.github.AdvertflairPublished by GitHub account Advertflairv1.0.0Updated Nov 3, 2026
Conversation drift visualization for Claude. Color-coded inflection points and one-click forking.
Overview
Tracks drift in Claude Code conversations, scoring each message against the original intent and letting you fork a clean branch from any point.
- What you can do
- Track messages in a named session and get a drift score with a color code
- View a color-coded timeline map of the whole conversation
- Fork from any message index into a named branch
- Extract a clean context packet for a branch, e.g. for handoffs
- Check current drift status, archive and reset a thread, and show the color legend
- When to use it
- Useful when long Claude Code sessions drift away from the original goal and you want to spot the pivot point and restart from a clean context. Also handy for summarizing decisions before a team handoff.
- Requirements
- Runs locally over stdio via npx threadmap-mcp, so Node.js and npx are needed. No accounts, API keys, environment variables or headers are declared. The README describes adding it to Claude Code or a Claude desktop config file.
Installation
In SourceWeft
- Open Threadmap Mcp in the dashboard and add it to a workspace.
- Enable the server for the chats that should use its tools.
Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.
Other MCP clients
Follow the launch instructions in the repository.
README
🗺️ ThreadMap MCP for Claude
Conversation drift visualization and forking for Claude.
Stop losing your original intent. See exactly where your conversation drifted. Fork from any point and pick up with clean context.
See it in action
Color dots on every message — live drift tracking:
Timeline panel with fork buttons at every inflection point:
Two ways to use ThreadMap
Both are in this repo. Use one or both.
Chrome Extension — Visual dots in claude.ai
The Chrome extension injects ThreadMap directly into the claude.ai interface. No config. Just install and open Claude.
Install (2 minutes)
What you see
- Color dot next to every message showing drift % from your original intent
- Sticky bar at the top with live drift status and message count
- ↺ rescan — loads all messages from long previous conversations
- timeline — full conversation map, click any row to jump to that message
- ⑂ fork — opens a new Claude chat with full context pre-loaded in the input
How forking works
- Click ⑂ fork in the timeline or next to any message bubble
- A new claude.ai tab opens automatically
- The full context (origin intent + all messages up to that point) is pasted into the input
- Review and hit Send — Claude picks up from the clean version
Chrome extension files
Heads up: The extension uses Developer mode (load unpacked). A Chrome Web Store version is on the roadmap.
MCP Server — Claude Code integration
The MCP server adds 7 tools to Claude Code for programmatic drift tracking.
Install
Or manually add to ~/.claude/claude_desktop_config.json:
Quick start
All 7 tools
Color system
Orange and Red = fork candidates. The conversation has left your original intent behind.
Real-world use cases
Claude Code — architecture integrity Track every message while building. Orange = scope creep. Fork before the drift gets worse. Architecture decisions from the first 10 messages stay clean.
Long previous conversations Click ↺ rescan in the Chrome extension. ThreadMap scrolls to the top, loads all messages, scans every one, then returns you to the bottom — all dots appear.
Team handoffs
Use threadmap_branch_context to extract a clean summary of what was decided. Paste it into a new session — no scrolling through 200 messages.
Why open source?
Non-linear conversation navigation is the missing primitive for AI-native work. Right now Claude conversations are books with no table of contents, no chapters, no bookmarks.
ThreadMap is the table of contents. If this becomes the standard, it travels across Claude, GPT, Gemini, anything. Nobody owns it. Everybody benefits.
Roadmap
- Local embedding model (MiniLM) for richer semantic scoring
- Chrome extension on Chrome Web Store (no developer mode required)
- Cross-session persistence (SQLite mode)
- Visual timeline export (SVG/HTML)
- GPT / Gemini compatibility
Contributing
MIT licensed. See CONTRIBUTING.md.
PRs welcome especially for: better drift scoring, Chrome extension improvements, and other AI platform support.
License
MIT — see LICENSE
Built by Advertflair. Open source. Ship it.
Source: README.md at commit 9bfa530
Tools
0Version history
1- v1.0.0LatestSep 16, 2026
Score
50% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.
- Reached in the last 30 days25 pts
Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.
- Auth declared10 pts
Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.
- Tools listed15 pts
Answer tools/list without a credential, or ship a package that starts in a sandbox without one.
- DescriptionRequired15 pts
- Logo10 pts
- Documentation15 pts
- License10 pts