Business Software for Anthropic

Scribe is an automated knowledge base that operates on a self-hosted platform, utilizing your tools to generate content. It analyzes the Git history, Claude Code and Codex sessions, as well as self-sent URLs and drop files, transforming this information into a well-organized, cross-project wiki formatted in Markdown and stored within Git repositories. By eliminating the need for developers to maintain an additional cognitive framework or to reconstruct context each time an agent session begins anew, Scribe effectively captures critical decisions, fixes, evaluations, and the rationale behind them, enabling agents to reference this accumulated knowledge prior to taking action. Its operational pipeline is scheduled via cron jobs, allowing it to identify projects and sift through low-value data using FTS5 before engaging with an LLM, thus extracting verified facts through structured bounded workflows that run in two passes. The final output is organized into entity-centric pages complete with YAML frontmatter, wikilinks, backlinks, retrieval context, and various typed relationships, including supersedes, contradicts, derived_from, specializes, and extends, ensuring comprehensive documentation and clarity across projects. This systematic approach significantly enhances collaboration and knowledge sharing among teams, streamlining the development process.

Mavvrik operates as a sophisticated platform for managing costs associated with AI and hybrid infrastructure, providing a centralized hub for finance, FinOps, IT, and engineering teams to oversee GenAI, autonomous agents, GPUs, cloud systems, on-premises resources, Kubernetes, data platforms, and SaaS solutions. By consolidating cost, usage, and telemetry data from major providers such as AWS, Azure, Google Cloud, Oracle, VMware, NVIDIA, OpenAI, Anthropic, Gemini, Snowflake, Databricks, and LiteLLM, it establishes a comprehensive source of truth for the entire technology ecosystem. Teams can meticulously monitor each model interaction, agent engagement, GPU utilization, and resource workload, allowing for precise spending allocation across various dimensions, including customer, product, feature, project, application, environment, team, or cost center. Through in-depth analysis of cost-to-serve and unit economics, Mavvrik uncovers margin losses, identifies costly workloads, and clarifies the actual expenses involved in delivering each service. Additionally, its capability for real-time anomaly detection and alerts serves to flag unusual usage patterns before they escalate into unexpected budget overruns, while its predictive forecasting tools assist organizations in effectively modeling their cloud, GPU, and AI-related expenditures. This holistic approach empowers teams to make informed financial decisions and optimize resource utilization for sustained growth.

Router acts as a gateway designed to lower inference costs by selecting the most cost-effective model that satisfies performance requirements for each request. It simplifies access for developers by providing a single endpoint and API key, allowing them to utilize a variety of both closed and open-source AI models from numerous providers, including OpenAI, Anthropic, Grok, and Fireworks, thereby eliminating the need to connect to each provider individually. Initially, requests are processed through Router, which enables tracking of usage, model selection, provider information, and associated costs, ensuring that workloads are efficiently directed to alternative options when quality remains intact. With Router Strategies, developers can establish their own cost and performance priorities for different request types or rely on pre-set benchmarks derived from actual production experiences. The system is responsive to real-time conditions such as latency, availability, failures, and rate limits, allowing for the seamless rerouting of eligible requests to other available models when a particular provider is unable to fulfill them. This flexibility enhances the overall efficiency and reliability of the service, ensuring that developers can meet their application demands effectively.

Onyx serves as a robust AI control platform designed for the discovery, protection, governance, optimization, and evaluation of AI agents and models within an organization. It provides crucial visibility for security, governance, and AI teams into both authorized and unauthorized AI activities across various environments, including SaaS applications, cloud services, endpoints, and code, encompassing aspects such as prompts, responses, and agent behaviors. The AI Security feature enhances the organization's security posture by pinpointing vulnerabilities and implementing real-time safeguards against potential threats and misuse. Meanwhile, AI Governance ensures compliance with security standards and regulatory mandates by offering opt-in coverage and allowing policy controls to be articulated in natural language. Additionally, AI Orchestration streamlines the process of deploying agents and Multi-Cloud Platforms (MCPs), while optimizing for cost, accuracy, and latency. The AI ROI component facilitates the measurement of adoption, the establishment of objectives, and the tracking of results across various departments within the organization. Furthermore, the Onyx Guardian Agent functions as an overseeing AI, perpetually identifying risks and resolving issues throughout the platform, thereby enabling organizations to effectively manage a large number of agents seamlessly. Ultimately, Onyx empowers businesses to harness the full potential of AI while maintaining control and oversight.

Kana is an innovative marketing platform that empowers teams to convert disjointed marketing data into timely decisions and quantifiable business results. By integrating data throughout the marketing landscape, it consolidates information that is typically scattered across various tools and uncovers emerging patterns, risks, behavioral cues, and opportunities. The platform provides recommendations and implements next-best actions, eliminating the need for manual speculations with decisions grounded in data aimed at enhancing performance. Serving as a marketing operating layer, Kana interfaces with email services, advertising platforms, CRMs, spreadsheets, and other software, utilizing AI agents to synchronize data and actions seamlessly. Unlike traditional generative AI that merely responds to user queries, its agents are designed to pursue objectives, orchestrate complex workflows, analyze signals in real-time, and carry out actions throughout the marketing infrastructure, thus enabling a more cohesive and efficient marketing strategy. This holistic approach ensures that teams can navigate their marketing efforts with greater agility and insight.

oMLX is an MLX server specifically designed for macOS, enhancing the efficiency and speed of local AI operations on Apple Silicon. It caters to the functional dynamics of coding agents by implementing paged SSD KV caching, which enables the persistence of cache blocks on disk; this means that previously accessed prefixes can be retrieved quickly across different requests and even after server restarts, thereby eliminating the need to recompute them from scratch. As a result, the time taken to generate the first token in lengthy contexts can be significantly reduced, dropping from a range of 30 to 90 seconds down to less than five seconds after the initial interaction. The server adeptly manages simultaneous requests through a continuous batching mechanism via mlx-lm’s BatchGenerator, which enhances overall generation throughput without requiring requests to queue up behind a single task. oMLX is capable of simultaneously serving a variety of models, including LLMs, vision-language models, embedding models, and rerankers, utilizing LRU eviction to manage memory constraints effectively. Furthermore, it is compatible with any MLX-format model sourced from Hugging Face, such as Qwen, LLaMA, Mistral, Gemma, DeepSeek, MiniMax, and GLM, and can also utilize models that are already present in the standard Hugging Face cache, directories associated with LM Studio, or any custom storage locations, ensuring a versatile user experience. This flexibility in model integration enhances the overall usability and practicality of oMLX for developers and researchers alike.

Open Code Review is a command-line interface for code evaluation powered by artificial intelligence, originating from Alibaba's robust internal code review tool and tested through millions of practical applications. It analyzes Git diffs, communicates with a configurable language model via an agent capable of utilizing various tools, and produces organized review feedback with exact line references. Instead of limiting itself to the visible differences, the agent has the ability to access entire files, browse the codebase, examine related modifications, and cross-check context, which allows it to identify more significant issues. Its hybrid structure distinguishes between deterministic engineering functions—such as filtering files, dividing tasks, routing rules, scheduling, and positioning lines—and the reasoning of the language model for detecting risks, exploring context, and classifying problems. Additionally, a specialized reflection module is included to catch any inaccuracies or knowledge gaps before the feedback is delivered. Users can modify the review depth, adjusting it from minimal to extensive based on their need for either rapid assessments or comprehensive evaluations. This flexibility ensures that developers can tailor the review process to their specific project demands.

AstrBot is a versatile, open-source AI assistant and chatbot platform designed for diverse applications in messaging, automation, and collaborative efforts. Operating seamlessly across various messaging applications, it features an agent runtime that accommodates Sub-Agents, intricate workflows, tool interactions, context management, scheduled tasks, runtime controls, and secure agent environments for executing real-world actions. The platform's integration of native MCP and Skills enhances its AI functionalities, while a plugin-driven framework grants users access to over 1,000 community-created extensions aimed at boosting productivity, facilitating group activities, managing content workflows, and integrating with external systems. Supporting a variety of AI providers, including OpenAI, Google, and Anthropic, AstrBot offers flexible model switching alongside capabilities such as text generation, visual processing, speech recognition, speech synthesis, and embedding functions. Additionally, it comes equipped with a comprehensive knowledge base that enables parsing of PDF, DOCX, and Markdown files, employs hybrid dense and BM25 retrieval methods, and allows referencing multiple knowledge bases within a single dialogue, thereby enriching user interactions. Overall, AstrBot stands out as an innovative solution for enhancing communication and productivity across a multitude of environments.

Opal Zero serves as a governance platform for AI agents that provides just-in-time access management by cataloging every agent and linking it with a responsible owner, evaluating access requests, and implementing decisions through the existing MCP gateway. This comprehensive solution allows organizations to maintain a unified inventory across various identity providers and AI platforms, including Okta, Entra, Anthropic, OpenAI, Bedrock AgentCore, and Cursor, revealing the permissions each agent holds and their respective ownership. The Risk Center highlights access that may be misaligned with intended use, unassigned, or excessive, directing concerns to the relevant owner while offering immediate corrective measures. Paladin assesses requests against established policies and contextual information, pinpoints the least-privileged access routes, adheres to ownership boundaries, and elucidates the rationale behind each resolution. By analyzing real user behavior, Policy Insights uncovers instances of redundant access and suggests enhancements to access policies, moving beyond sporadic fixes. Furthermore, Gateway Enforcement ensures that approved access decisions are implemented as specific, temporary policies within the existing gateway infrastructure, thereby fostering a more secure and organized environment for managing AI agents.

BugIt serves as a locally operated quality assurance tool that transforms rough test notes into organized bug reports tailored for software development teams. It prompts users for any missing information, analyzes logs, investigates connected tracking systems for potential duplicates, proposes severity levels, redacts sensitive information, and ensures the completeness of each report. Users are required to review every draft and type FILE IT before BugIt can create or modify any tickets. The tool is compatible with a variety of platforms, including Jira, Azure DevOps, GitHub Issues, GitLab Issues, Bugzilla, YouTrack, Linear, Shortcut, ClickUp, Asana, and Trello, and can be operated via compatible AI assistants or through a terminal interface. BugIt aims to assist testers, developers, and teams in generating consistent and actionable reports while safeguarding tracker credentials and project-related data on the user's device. Its intuitive design ensures that team members can easily communicate and resolve issues without compromising security.

Caterfli serves as a hosted platform for AI agents, facilitating connections to PostgreSQL or MySQL databases, REST APIs, OpenAPI services, and MCP servers. It identifies existing tables, endpoints, and tools, subsequently creating draft agents for user review and customization. Teams can interact using natural language via web chat, with additional access through Telegram, Slack, and WhatsApp available on certain plans. These agents possess the capability to read data, create or modify records, generate reports, and delegate tasks to other agents as needed. Workspace roles and permissions are established to restrict actions based on user authority, while policy checks ensure that each tool call and its parameters are validated. Actions deemed destructive or high-risk are designed to require human intervention before execution. Comprehensive execution traces and change logs facilitate thorough review processes. Caterfli caters to teams desiring a conversational interface for their existing business systems without the need to transfer their data elsewhere. The free plan offers one connection alongside one agent, while premium options provide enhanced limits and additional functionalities for users. Ultimately, Caterfli empowers teams to leverage AI in a seamless manner, enhancing their operational efficiency.

Polyant is a collaborative open-source platform designed to assemble teams of specialized AI agents that assist company personnel by managing tasks such as sales briefs, customer triage, document extraction, and compliance reporting. With its modular and self-hosted framework, teams have the flexibility to select their preferred model providers, including open-weight models, and can utilize their own infrastructure, ensuring they are never tied to a single vendor. From the outset, the platform includes governance and auditing tools tailored to meet EU AI Act requirements, with all agent decisions meticulously recorded in a compliance-friendly audit registry accessible to compliance teams. Organizations maintain comprehensive control over data residency options, allowing deployment on-premises, within their own cloud, or utilizing EU infrastructure managed by Exelab. Furthermore, the deployment of agents can occur in mere days rather than the typical months, all while retaining self-hosting capabilities and adhering to governance standards. This solution is particularly tailored for mid-to-large European enterprises operating in regulated industries such as banking, insurance, and healthcare, where relying solely on a hosted API is not a viable option, ensuring that companies can effectively manage their AI integration in a compliant manner.

WebTwizz is an AI-driven full-stack application builder designed to help users create complete web apps through simple natural language instructions. The platform allows users to build websites, dashboards, booking systems, online stores, admin panels, portfolios, blogs, and SaaS applications without needing traditional development workflows. Its AI assistant can generate interfaces, add features, debug issues, and refine projects through conversational interaction, making app creation faster and more accessible. WebTwizz combines a visual drag-and-drop editor with advanced customization tools for typography, layouts, spacing, colors, and reusable components. The platform also includes a built-in content management system and database integration capabilities for managing dynamic content such as products, users, blog posts, and customer data. Users can organize pages, structure app layers, and create reusable custom components to streamline development and maintain design consistency. WebTwizz integrates with major third-party services including Stripe for payments, Supabase for databases and authentication, OpenAI and Anthropic for AI features, and PostHog for analytics. The platform supports responsive web app creation while offering AI-powered workflows that automate much of the design and development process. WebTwizz is designed for both technical and non-technical users who want to create professional-grade applications quickly and efficiently. Its flexible pricing plans support personal projects, startups, agencies, and larger development teams with scalable publishing and AI credit options. By combining no-code tools with AI-assisted development, WebTwizz helps users build and launch full-stack applications faster than traditional development methods.

Versionveil serves as a real-time intelligence platform focused on vendor changes specifically designed for engineering teams. It keeps a close watch on various vendors such as OpenAI, Stripe, Vercel, Supabase, Anthropic, and Cloudflare by tracking modifications in their APIs, pricing, SDKs, and infrastructure, which are often dispersed across multiple changelogs, documentation, and status pages. By transforming these updates into structured alerts that include severity ratings, concise summaries, and AI-generated analyses of the impact—detailing what has changed and why it is significant—Versionveil enhances communication and awareness among teams. These alerts are swiftly delivered via Slack, Discord, or email, ensuring that the relevant teams are promptly informed about important changes, while all updates are archived in a searchable history of vendor modifications. This functionality not only mitigates dependency risks but also helps prevent unexpected production issues arising from third-party changes, ultimately fostering a more resilient engineering environment.

OfoxAI serves as a comprehensive API gateway compatible with OpenAI, allowing developers and teams to seamlessly access over 100 large language models—including GPT, Claude, Gemini, and DeepSeek—through a single endpoint and one API key. Say goodbye to the hassle of managing multiple accounts, SDKs, and invoices: with OfoxAI, you can integrate once, switch between models with ease, and expand from a single prototype to a full-fledged production team effortlessly. Key features include: One API Key, Access to 100+ Models — Stay current with the latest offerings from OpenAI, Anthropic, Google, DeepSeek, and others. Three Native Protocols — Full compatibility with OpenAI, Anthropic, and Gemini SDKs, enabling seamless transitions without code alteration—just change the base URL. Low-Latency Access — Benefit from global routing with an average latency of under 300ms for quick response times. Zero Markup Pricing — Enjoy transparent pricing, paying only the standard rates set by the official providers, free from hidden fees or surcharges. Built for Teams — Utilize a shared billing dashboard, track usage by each member, and implement budget controls effectively. Flexible Payment Options — OfoxAI accommodates various payment methods, including credit cards, PayPal, and other major regional options for convenience and accessibility. Plus, its user-friendly interface ensures that teams of all sizes can navigate the platform with ease.