SessionSift

Toolsgithub.com

SessionSift

Local-first analysis of Codex and Claude Code sessions: dashboard, comparisons and ten read-only MCP tools. No account or cloud. - LanceSandino/sessionsift

Quick read

A local-first tool for analyzing Codex and Claude Code sessions with a dashboard, comparisons and ten read-only MCP tools. Use it to inspect, compare and manage conversation transcripts locally without accounts or cloud services.

  • Browse and inspect Codex and Claude Code conversation tokens
  • Compare sessions and run ten read-only local MCP tools
  • Import/export transcripts and run a local web dashboard

Reactions

No reactions yet. Record a short about this link and it shows up here, in Reels and in the community feed.

Is github.com your site? Sign in to verify it and get a Verified badge, full stats and a link-back badge.

A closer look at SessionSift

What SessionSift is

SessionSift is a local-first desktop/web tool for analyzing conversation transcripts from Codex and Claude Code and for explicitly imported ChatGPT/Claude exports. It provides a dashboard, session comparisons and ten read-only MCP tools, and is distributed from the project repository with release artifacts and checksums. The project is source-available under the PolyForm Noncommercial 1.0.0 license (not OSI open source).

What you can actually do with it

Run a local service that serves a dashboard on http://127.0.0.1:7777 to browse conversations, inspect tokens and tools, compare sessions, and use the read-only MCP tools. You can install or build platform binaries (macOS Intel and Linux amd64 are supported in the listed release), or run with Docker using the provided Compose instructions; the app stores data in ~/.sessionsift. The local guide and docs explain import/export, Conversation/Timeline/Tools/Review flows, MCP preview/configuration/undo, backups and uninstall.

Who will find SessionSift useful

People who need local, read-only analysis of Codex and Claude Code sessions and who want to keep data off cloud services will get the intended experience, including users comfortable running local binaries, Homebrew formulas, or Docker builds. It is aimed at users who will explicitly enable MCP hooks and who will follow the security guidance (do not expose the dashboard to a network and report issues privately).

Questions people ask about SessionSift

Do I need an account or cloud service to use SessionSift?
No account or cloud service is required; the tool is local-first and runs as a user service or via Docker on your machine.
Which platforms does the release support?
The documented release supports macOS Intel (amd64) and Linux amd64 only; Apple Silicon and ARM builds are not offered in that release.
Will SessionSift send my transcripts to a third party?
The project emphasizes local use and notes that MCP hooks are optional and disabled until enabled; however it also warns that connected AI clients may send tool output to their own provider, so local MCP does not guarantee client offline behavior.

SessionSift was shared on NexURLs by NexURLs on November 7, 2026, in Tools.

What we found on this site

Google safety check
No threats found

From public records. See how the rest of the web compares in the NexURLs Internet Report.

Quick answers

What is SessionSift?

SessionSift (github.com): A local-first tool for analyzing Codex and Claude Code sessions with a dashboard, comparisons and ten read-only MCP tools. Use it to inspect, compare and manage conversation transcripts locally without accounts or cloud services.

Who shared SessionSift on NexURLs?

It was shared by NexURLs in November 2026. Every link on NexURLs is shared by a real member and passes a safety check before it goes out.

What are some sites like SessionSift?

Members found these close to it: Local-First Flight Recorder for AI Coding Agents, Open-Source Memory System for AI Coding Agents, Title. The full list is on this page under Sites like SessionSift.

Sites like SessionSift

More Tools finds

All Tools websites →

Feeling lucky? Try the random website generator.

Where people went next

Nobody has added a next stop yet. Be the first.

Five finds like this, every morning

Five great websites every morning, picked by people and written up by AI. Free, and one click to stop.

Remember Me
Scroll to Top

More

📰 Latest Posts
📊 Community Stats
👋 New Members
👥 Popular Groups