claude-code - 💡(How to fix) Fix Feature request: Import or continue claude.ai conversations in Claude Code [1 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
anthropics/claude-code#48990Fetched 2026-04-17 08:53:51
View on GitHub
Comments
0
Participants
1
Timeline
2
Reactions
0
Author
Participants
Timeline (top)
labeled ×2

Root Cause

It is genuinely strange that Claude Code can read my Slack DMs and Notion pages but cannot read the conversation I had with Claude itself thirty minutes ago. Closing this gap would make Claude feel like a single continuous assistant across surfaces instead of two disconnected products that happen to share a name.

Fix Action

Fix / Workaround

Current workaround

RAW_BUFFERClick to expand / collapse

Problem

There is no way to bring a conversation from claude.ai (web/desktop) into Claude Code. The two products share the same underlying model and company, but their conversation histories are completely siloed.

A very common workflow is:

  1. Brainstorm or scope out an idea with Claude on claude.ai (easier for long-form discussion, mobile, etc.).
  2. Move to Claude Code to actually implement it.

Today the only way to carry context across is to ask claude.ai to summarize the discussion into a markdown blob and paste it into Claude Code. That works, but it feels like the wrong answer within the same Claude platform — especially when third-party MCP integrations (Slack, Notion, Gmail, Google Drive, etc.) can pull in external conversations, yet Claude's own conversation history cannot.

Current workaround

  • Ask claude.ai to produce a markdown summary / brief
  • Copy-paste it into Claude Code as initial context
  • Lose nuance, intermediate reasoning, and attachments along the way

Proposed solutions (any of these would help)

  1. First-party claude.ai MCP server — search and fetch past claude.ai conversations from inside Claude Code, the same way the Slack/Notion MCPs work.
  2. claude --import <conversation-id> / "Continue in Claude Code" button — a direct handoff from a claude.ai conversation into a new Claude Code session with the transcript as context.
  3. Structured export format — at minimum, make the claude.ai "Export data" output trivially consumable by Claude Code (e.g. claude --import-export ./export.zip).

Why this matters

It is genuinely strange that Claude Code can read my Slack DMs and Notion pages but cannot read the conversation I had with Claude itself thirty minutes ago. Closing this gap would make Claude feel like a single continuous assistant across surfaces instead of two disconnected products that happen to share a name.

Environment

  • Claude Code CLI
  • claude.ai web + desktop

extent analysis

TL;DR

Implementing a first-party claude.ai MCP server or a direct handoff feature from claude.ai to Claude Code would likely resolve the issue of siloed conversation histories.

Guidance

  • Investigate the feasibility of integrating a first-party claude.ai MCP server to enable searching and fetching past conversations within Claude Code.
  • Consider adding a "Continue in Claude Code" button to claude.ai conversations, allowing for a direct handoff of conversation transcripts to Claude Code.
  • Explore the possibility of creating a structured export format for claude.ai conversations that can be easily imported into Claude Code using a command like claude --import-export ./export.zip.
  • Evaluate the potential benefits and challenges of each proposed solution, including technical requirements, user experience, and data consistency.

Example

No code snippet is provided as the issue focuses on high-level integration and feature requests rather than specific code implementation.

Notes

The solution may require significant development and infrastructure changes, and it is essential to consider factors like data privacy, security, and consistency across different platforms.

Recommendation

Apply a workaround, such as the proposed first-party claude.ai MCP server or direct handoff feature, as it would provide a more seamless user experience and address the core issue of siloed conversation histories.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING