Anthropic Unveils Ten Finance-Focused AI Agents, Integrates Claude into Microsoft Office
Key Points
- Anthropic launched ten AI agents built for banks, insurers and other financial firms.
- Agents run on Claude Opus 4.7, which outperformed GPT‑5.5 and Gemini 3.1 Pro on a finance benchmark.
- Each agent combines skills, connectors and subagents for modular, customizable workflows.
- Five agents support research and client coverage; five focus on finance and operations.
- Connectors provide real‑time access to data from FactSet, S&P Capital IQ, MSCI, PitchBook, Morningstar, Chronograph, LSEG, Daloopa and more.
- Claude is now integrated directly into Microsoft Excel and other Office apps.
- Agents are available across paid Claude Cowork and Claude Code plans; Managed Agents remain in public beta.
- Deployment timeline advertised as days rather than months.
- Financial institutions can tweak agents to meet strict regulatory standards.
Anthropic announced the launch of ten pre‑built AI agents tailored for banks, insurers and other financial firms. The agents, powered by the new Claude Opus 4.7 model, are available through Claude Cowork, Claude Code and a Managed Agents beta, and can be deployed in days rather than months. Anthropic also highlighted Claude’s new integration directly into Microsoft Excel and other Office apps, promising faster, more accurate handling of time‑consuming financial tasks.
Anthropic rolled out a suite of ten finance‑specific artificial‑intelligence agents designed to streamline some of the most labor‑intensive processes in banking, insurance and related sectors. The new agents, built on the Claude Opus 4.7 model, are accessible via the company’s Claude Cowork, Claude Code and the Managed Agents program, which remains in public beta.
Each agent combines three components—skills, connectors and subagents—to form a modular workflow. Skills provide the instruction set, connectors pull data from external sources, and subagents act as auxiliary Claude models called on for specialized tasks such as comparable selection or methodology checks. This architecture lets financial institutions tailor the agents to meet strict regulatory requirements.
Anthropic positioned the offering as a rapid‑deployment solution, promising implementation in “days rather than months.” Five of the agents focus on research and client coverage, while the other five target finance and operations. Notable examples include a pitch‑builder agent that assembles client presentations, a market‑researcher agent that scans industry data, and a valuation‑reviewer agent that cross‑checks financial models.
Claude Opus 4.7, the engine behind the agents and the new Excel connectors, earned a 64.37% score on Vals AI’s Finance Agent benchmark—well ahead of competitors GPT‑5.5 (59.96%) and Gemini 3.1 Pro (59.72%). Anthropic described the model as “state‑of‑the‑art on financial tasks,” underscoring its edge in handling complex calculations and data‑intensive queries.
The agents can tap into a range of market‑data and research platforms through built‑in connectors. Sources include FactSet, S&P Capital IQ, MSCI, PitchBook, Morningstar, Chronograph, LSEG and Daloopa, among dozens of others. Real‑time data flows directly into the agents, enabling up‑to‑the‑minute analysis without manual data entry.
In parallel with the agent launch, Anthropic reminded users that Claude is now embedded within Microsoft Excel and other Office applications. The integration lets finance professionals invoke Claude’s capabilities from within familiar spreadsheets, turning routine calculations and report generation into conversational interactions.
All paid Claude Cowork and Claude Code plans include access to the new finance agents, while the Managed Agents offering remains in a public beta for programmatic use. Companies can further refine the agents, adjusting skills or adding custom connectors to align with internal policies or proprietary data sets.
Industry observers see the move as a strategic push to capture a growing market for AI‑driven productivity tools in heavily regulated sectors. By bundling high‑performance models, extensive data hookups and a quick‑deployment framework, Anthropic aims to reduce the friction that typically slows AI adoption in finance.
Anthropic’s announcement arrives as financial firms increasingly seek automation to cut costs and accelerate decision‑making. The company’s emphasis on speed, customization and regulatory compliance positions its new agents as a compelling option for organizations looking to modernize legacy workflows without extensive engineering effort.