Frontier TPRM Agents Adds Monitoring Function, Both DD and Monitoring Functions Now Available as Claude Cowork Plugins

**[Overview]** Frontier Advisory & Consulting Inc. (Headquarters: Chuo-ku, Tokyo; President and CEO: Yohei Shoda; hereinafter referred to as "the Company"), a professional firm specializing in digital governance, has newly added a monitoring function to its AI agent suite "Frontier TPRM Agents," which supports the automation of Third-Party Risk Management (TPRM). We are pleased to announce that both the conventional Due Diligence (DD) function and the newly added monitoring function are now available as Plugins operating on Anthropic's AI agent platform "Claude Cowork." This initiative establishes a system where AI agents cover the entire TPRM lifecycle, from due diligence at the start of a contract to continuous monitoring during the contract period. Customers can start using it simply by installing the Plugin in their Cowork environment, significantly reducing the burden of environmental setup during implementation. **■ Enhancement of Monitoring Function** In TPRM, continuous understanding of supplier conditions during the contract period is considered important, not just the initial screening (DD) at the time of contract signing. However, in practice, designing evaluation frequencies, collecting and analyzing evaluation data, and understanding the cross-sectional status of multiple suppliers requires significant effort, posing a challenge for those in charge. The newly added monitoring function automates the continuous understanding of individual supplier conditions and the prioritization across multiple suppliers at frequencies corresponding to their importance (Tier). **■ Monitoring Function Configuration** UC | Name | Overview ---|---|--- UC-07 | Change Detection | Periodically detects changes in public information such as supplier finances, management, and incidents. UC-08 | SLA Evaluation | Compares SLA reports with KPI benchmarks to detect deviations and analyze trends. UC-09 | Regular Re-evaluation | Periodically re-evaluates management systems and analyzes differences from the previous evaluation. UC-10 | Incident Detection/Impact Assessment | Detects incidents and assesses their impact through report analysis. UC-17 | Cross-sectional Monitoring | Batch execution for multiple suppliers, prioritization, and cross-sectional view. Each UC is executed at a frequency corresponding to the supplier's importance (Tier 1/2/3), and the collected results are consolidated in cross-sectional monitoring (UC-17). Managers and executives can proceed with addressing high-risk suppliers through prioritized alerts and overall status. **■ Provided as Claude Cowork Plugin** Both the DD function and the monitoring function are now offered as Plugins operating on Anthropic's AI agent platform "Claude Cowork." By offering it as a Plugin, we provide the following benefits to customers: - You can start using it in a short period simply by installing it in your Cowork environment. - Your unique data is processed securely within your Cowork environment. - Business users can operate it in a chat format, requiring no command-line or programming knowledge. - It can be operated in combination with Cowork's standard access control and audit log functions. The following two types of Plugins are provided: Plugin Name | Purpose | Included Functions ---|---|--- tprm-due-diligence | Due diligence at contract signing | Supplier company evaluation, importance rating, market evaluation, management system evaluation, gap analysis/improvement recommendations, comprehensive DD report. tprm-monitoring | Monitoring during contract period | Change detection, SLA evaluation, regular re-evaluation, incident detection/impact assessment, cross-sectional monitoring. **■ Overview of Frontier TPRM Agents** "Frontier TPRM Agents" is a solution that automates each TPRM task with AI agents. From supplier company evaluation to ongoing monitoring during the contract, it significantly reduces practical man-hours while supporting decision-making by those in charge. **** Human-in-the-Loop: Incorporates a flow where humans always confirm and approve critical judgment points, designed not to blindly trust AI judgments. Complete Traceability: Records all judgment grounds, information sources, and execution history, structured to fulfill audit requirements and accountability. Sub-agent Review: Results of main processes are independently reviewed by another AI agent, ensuring judgment quality and consistency. **** Efficiency: Reduces man-hours compared to manual consultant work. Homogenization of Quality: Uniform application of evaluation standards and structural prevention of omissions and errors. Shortened Lead Time: From start to completion of work.