How to Migrate from Pocket to ContextBolt (2026 Guide)

Pocket is gone. Mozilla shut the service down on July 8, 2025 and finished deleting user data on November 12, 2025 after two extensions of the original export deadline. If you exported your library in time, you have a CSV sitting on a hard drive somewhere. If you did not, your years of saves are not coming back.

Either way, the migration question is the same. Where does this stuff actually live now? The honest answer is “in two places, not one.” Pocket did two jobs that nobody has fully replaced under one roof: it saved articles for distraction-free reading, and it was a generic save-anything bucket that ended up holding a lot of social content the reading view never handled well.

This guide walks the full migration in about 30 minutes. It covers the article half (Raindrop or Instapaper, with the Pocket CSV import) and the social half (ContextBolt, which captures X, Reddit, and LinkedIn saves natively and makes them searchable by meaning). Use what fits. Skip what does not.

What “migrate from pocket” actually means in 2026

There is a category error baked into most Pocket migration guides. They treat Pocket as if it were a single product with a single replacement. It was not. Pocket was a save-anything bucket pretending to be a read-later app.

Look at your old Pocket library, if you still have the CSV, and the pattern is almost always the same. A long tail of articles you genuinely wanted to read later. A bigger pile of tweets and Reddit links you bookmarked because the headline was interesting. A heap of YouTube videos. A surprising number of LinkedIn posts. The reading view was excellent for the article half. For everything else, Pocket stripped the context that made the save useful in the first place. A tweet without the thread is half a tweet. A Reddit post without the comments is missing the punchline.

The 2026 migration is not “find a single Pocket clone.” It is “split your old library along the lines Pocket was hiding.”

• Articles go to a dedicated read-later app. Raindrop or Instapaper. Both import Pocket exports cleanly.
• Social saves (X, Reddit, LinkedIn) go to a tool that captures the actual post, the thread, the comments. ContextBolt is the only one built for this.
• Videos and assorted links can sit alongside articles in Raindrop, which is generic enough to handle them.

The two-tool stack is the migration. Trying to force everything into one bucket is what made Pocket frustrating to use in the first place.

Before you migrate: figure out what you actually saved

Five minutes of triage saves an hour of regret later. Open your Pocket CSV in any spreadsheet app and sort by the URL column. The shape of the library tells you which half matters most.

• Mostly articles (Medium, Substack, blogs, news sites). The article half is where most of your value lives. Send the CSV to Raindrop or Instapaper. ContextBolt is still useful for new social saves, but the migration is article-led.
• Mostly social (twitter.com, x.com, reddit.com, linkedin.com URLs). The article import will feel thin because Pocket stripped most of the value out of those saves anyway. The bigger win is ContextBolt going forward, where the full post and thread get captured natively.
• A genuine mix. Run both. The two tools are designed to coexist and the total setup cost is still under 30 minutes.

If you missed the export window and have no CSV, skip the triage. The forward-looking stack is the same and the steps below still apply. You just lose the historical articles.

Step 1: install ContextBolt for the social half (3 minutes)

Start with the social half because it is the one that has zero CSV-import friction. There is nothing to upload, nothing to map, nothing to clean.

1. Open the ContextBolt Chrome listing and click Add to Chrome.
2. Pin the extension so the icon sits in your toolbar. You will not interact with it much, but it needs to be installed.
3. Open X (or twitter.com). The extension picks up your bookmarks automatically through a fetch interceptor. You do not click anything. New saves from this point are captured in real time.
4. Open reddit.com. The DOM observer starts watching your saves the same way.
5. For LinkedIn, you have two paths. Either click save on a post and ContextBolt captures it going forward, or use LinkedIn’s official data export to bulk-import your historical saves in one batch. The bulk import is the fastest way to backfill a long save history.

That is the full social migration. No CSV, no field mapping, no platform-by-platform manual transfer. AI tags every save with a main topic and 2 to 4 specific tags. Topics auto-cluster in the sidebar for filtering. Semantic search means you can type what you remember instead of what you saved. If you want to understand how semantic search beats keyword search for saved posts, the semantic search guide goes deeper.

The free tier covers 150 saves across all three platforms with AI tagging and search. Pro is $6 a month for unlimited saves, encrypted cloud sync, and a personal MCP endpoint that connects to Claude Desktop, Claude Code, Cursor, or Windsurf. Most people start free. The Pro decision shows up the first time you want to ask Claude about your saves from inside a chat.

Step 2: import your Pocket CSV to Raindrop or Instapaper (10 minutes)

This is the half that uses your old Pocket export. If you missed the November 12 deadline, skip this section and start using Raindrop or Instapaper for new article saves only.

Raindrop (closest free Pocket replacement)

Raindrop.io is the closest free one-to-one Pocket clone for general bookmarking. Unlimited bookmarks on the free tier, full-text search, browser extensions, iOS and Android apps, decent reader view. Many ex-Pocket users land here and stop.

1. Sign up for a free Raindrop account at raindrop.io.
2. Open Settings, then Import.
3. Upload your ril_export.html file from Pocket. Raindrop accepts the HTML export directly. CSV works too, plus JSON, TXT, and ENEX if you exported in a different format.
4. Wait for the import to finish. Large libraries (10,000+ items) take a few minutes.
5. Install the Raindrop browser extension. New article saves go here from now on.

What you keep: full URL list, titles, tags if you used them in Pocket. What you lose: the original “saved on” timestamps if your export was the basic HTML format, and any annotations Pocket stripped at export time.

Instapaper (cleanest read-later experience)

If reading is the point and you mostly used Pocket as a focused reading queue, Instapaper is a better fit than Raindrop. The text parser is still one of the cleanest on the market and the reading view is closer to what Pocket did at its best.

1. Sign up for a free Instapaper account.
2. Go to Settings, then Import.
3. Upload your Pocket CSV. The import takes about 5 minutes for most libraries.
4. Install the Instapaper browser extension. New article saves go here from now on.

What you keep: URL list, titles, basic metadata. What you lose: free-tier limits on search and highlights kick in fast. Premium is $3 a month and worth it if reading is the main use case.

The honest take on which to pick: Raindrop if you want generic bookmark storage with a reader bolted on. Instapaper if you want a reader with bookmark storage bolted on. Both are valid. Neither replaces the social half.

Step 3: stop trying to put everything in one app

This is the migration mistake almost everyone makes once. They install ContextBolt, like it, then try to use the share-to-Chrome menu to push an article URL into it. Nothing happens, because ContextBolt is not listening for that. Or they try to import their Pocket CSV into ContextBolt and get confused when the historical articles do not show up.

The two-tool stack works because each tool does one thing well. ContextBolt captures the full post content from X, Reddit, and LinkedIn (the thread, the author, the comments) and makes it searchable by meaning. Raindrop or Instapaper stores articles with a reading view. Forcing one tool to do both jobs is exactly what made Pocket frustrating for social content.

For most people the day-to-day after migration looks like this. You read an article, you hit the Raindrop or Instapaper button. You bookmark a tweet, ContextBolt picks it up automatically the next time you visit your X bookmarks page. You ask Claude “what was that thread about pricing strategy I saved last month?” and the MCP endpoint surfaces the actual tweet. You never think about which tool to use because the workflow routes itself.

If you want the deeper case for why bookmarks rot when you put them all in one folder, why you lose your best bookmarks covers the underlying behavior.

Migration mistakes to avoid

A few patterns show up over and over in ex-Pocket migration threads. They are easy to skip if you know to look.

• Importing the same library into three different tools “just to see.” You end up with three half-built second brains and never trust any of them. Pick the article tool and the social tool, then commit.
• Skipping the social half because “Pocket was fine for tweets.” Pocket was not fine for tweets. It saved a URL and stripped the content. Open any old Pocket save of an X thread and you will find a one-line title and nothing else.
• Paying for premium tiers on day one. Both Raindrop and Instapaper have generous free tiers. ContextBolt is free up to 150 saves. Use the free tier for a week before deciding whether the paid upgrade is worth it.
• Trying to maintain the old Pocket tagging system. Most Pocket tags were noise. Let AI tagging in ContextBolt and Raindrop’s auto-organization replace the manual taxonomy. You will not miss it.
• Forgetting LinkedIn. Half the migration guides act like LinkedIn saves do not exist. They do, and they are some of the most useful saves to make searchable. ContextBolt is currently the only tool that captures LinkedIn posts natively.

What changes after you migrate

The biggest change is psychological. Pocket trained a generation of knowledge workers to bookmark things and never look at them again, because the search was bad and the social context was missing. The two-tool stack inverts that pattern.

You stop saving things that look interesting and start saving things you might genuinely want to find later, because you can find them later. Semantic search closes the gap between “I remember saving something about this” and “here it is.” The reading view in Raindrop or Instapaper closes the gap between “I want to read this” and actually reading it.

The second change is that AI gets useful. Pocket was a black box from Claude’s perspective. ContextBolt Pro gives Claude an MCP endpoint that lets it search your X, Reddit, and LinkedIn saves mid-conversation. Ask Claude what you have saved about a topic and you get an answer that draws on your actual library, not a generic summary. That single capability is the reason most people upgrade to Pro within a few weeks of installing.

The stack you end up with

After 30 minutes of setup, the stack looks like this. ContextBolt handles X, Reddit, and LinkedIn. Raindrop or Instapaper handles articles. Both layers index automatically. New saves get tagged and made searchable without you organizing anything. The MCP endpoint (if you go Pro) makes the whole thing queryable from Claude.

It is not a one-to-one Pocket replacement. Nothing is, because Pocket was doing two jobs and one of them (the social-bookmark half) it never did particularly well. The two-tool stack is the honest answer. It costs $0 a month at the low end, $6 a month if you want the MCP layer, and $9 a month if you also pay for Instapaper Premium.

If you want the full comparison across every Pocket alternative (not just the two recommended here), the 8 best Pocket alternatives in 2026 breaks down where each one fits. For most people the answer ends at Raindrop plus ContextBolt or Instapaper plus ContextBolt. Two tools, 30 minutes, and your saves stop disappearing into a folder you never open.