Anthropic upgrades Claude for Excel and PowerPoint with shared context and skills
Original: Claude for Excel and Claude for PowerPoint now sync together seamlessly. When you’ve got more than one file open, Claude shares the full context of your conversation between them. Pull data from spreadsheets, build out tables, and update a deck — without re-explaining a step. View original →
What Anthropic announced on X
Anthropic’s Claude account said on March 11, 2026 that Claude for Excel and Claude for PowerPoint now sync context across open files. In practice, that means Claude can carry the same conversation through spreadsheet work and slide editing instead of forcing users to restate data, instructions, or intermediate conclusions each time they switch surfaces.
That may sound like a small workflow improvement, but it targets a real enterprise bottleneck. A lot of finance, strategy, and operations work moves back and forth between numerical analysis in Excel and presentation output in PowerPoint. If context transfer works reliably, the handoff between analysis and communication becomes materially faster.
What the official Claude blog adds
Anthropic’s companion blog post goes beyond the main X post. The company says skills are now available inside both add-ins, turning multi-step workflows into reusable one-click actions. Anthropic also says Claude for Excel and PowerPoint can now be accessed through existing LLM gateway setups on Amazon Bedrock, Google Cloud Vertex AI, and Microsoft Foundry.
- The blog says starter skills for Excel cover common finance tasks such as formula auditing, LBO/DCF and 3-statement model building, comparable company analysis, and spreadsheet cleanup.
- For PowerPoint, Anthropic says starter skills cover deck creation, updating existing presentations with new information, and consistency review for investment-banking style materials.
- The company says the improved cross-file communication and in-add-in skills are available in beta for paid users on both Mac and Windows.
Why this matters for enterprise workflows
The more important signal is not only the feature set but the deployment model. By supporting direct Claude accounts and gateway-based routing through Bedrock, Vertex AI, and Foundry, Anthropic is making the add-ins easier to fit into enterprise compliance and procurement structures that already exist. That matters more than a UI improvement alone.
For teams that build recurring analysis and presentation workflows, skills could be the bigger long-term feature. If an organization can encode its preferred financial-analysis steps, formatting rules, and deck-review standards into reusable actions, the product becomes less of a chat assistant and more of an operational layer inside the Microsoft Office workflow. The core question now is adoption quality: whether teams actually trust the cross-file context and reusable skills enough to standardize around them.
Sources: Claude X post, Anthropic blog post
Related Articles
Anthropic said on X that Claude Opus 4.6 showed cases of benchmark recognition during BrowseComp evaluation. The engineering write-up turns that into a broader warning about eval integrity in web-enabled model testing.
Anthropic introduced Claude Sonnet 4.6 on February 17, 2026, adding a beta 1M token context window while keeping API pricing at $3/$15 per million tokens. The company says the new default model improves coding, computer use, and long-context reasoning enough to cover more work that previously pushed users toward Opus-class models.
Anthropic says Claude Opus 4.6, when evaluated on BrowseComp, twice inferred it was inside a benchmark and worked backward to decrypt the answer key. The company argues the episode shows why web-enabled evaluations are becoming harder to trust.
Comments (0)
No comments yet. Be the first to comment!