v0.4.4

Updated Specification PDF to v0.4.4

The Agent BizOps Specification v0.4.4 introduces a major expansion of the framework by improving its structure and providing more comprehensive implementation guidance for enterprise AI business operations.

Key Changes:

• Renamed the document from Agent Metadata Specification to Agent BizOps to better represent its broader focus on enterprise AI business operations and governance.
• Expanded the Agent BizOps Framework with a hierarchical structure of stages and sub-steps, increasing the depth of implementation guidance.

v0.4.3

Updated Specification PDF to v0.4.3

The Agent BizOps Specification v0.5.0 introduces several enhancements that further expand the framework's coverage and implementation guidance for enterprise AI agents.

Key Changes:

• Renamed Agent Metadata Specification to Agent BizOps to better reflect the framework's broader focus on the operational lifecycle of AI agents.
• Expanded the framework from 25 to 38 sub-steps, providing more detailed guidance.
• Added mappings for additional sub-steps to the Tavro Agent BizOps open-source platform.
• Mapped the Data Governance sub-step to the Collibra How-to Guide available in the Tavro GitHub repository.

v0.4.2

Updated Specification PDF to v0.4.2

The Agent Metadata Specification v0.4.2 introduces Agent BizOps, a significant enhancement that expands the specification beyond agent metadata management into a comprehensive operational framework for enterprise AI agents. This new section establishes a structured Agent Lifecycle Framework that guides organizations through the complete journey of an AI agent, from initial planning and design to development, deployment, and ongoing monitoring. The framework includes key activities such as use case discovery, blueprint generation, stakeholder alignment, risk profiling, agent variant modeling, autonomy design, tool selection, prompt engineering, guardrail implementation, deployment planning, and security setup.

v0.4.1

Key updates in this version include:

• Agent Lifecycle – Introduced a new Agent Development Lifecycle covering the phases of Plan, Design, Develop, Deploy, and Monitor, providing a structured framework for managing AI agents across their operational journey.
• Agent BizOps – Added the new “From Agent Metadata to Agent BizOps” section, describing how Agent Metadata fuels Agent BizOps and supports enterprise AI agent adoption, governance, operational visibility, and lifecycle management.
• Contributing Guide – Updated the CONTRIBUTING.md document with the latest contribution guidelines and development workflow updates.
• Updated Specification PDF – All changes in this release are reflected in the updated Specification PDF.

v0.3.1

Key updates in this version include:

• MCP Server – Enhanced the MCP Server section with additional attributes and detailed examples for better clarity and implementation guidance.
• Agent Fairness – Introduced a new section covering agent fairness, including principles and considerations for equitable behavior in agent systems.

All changes in this release are reflected in the updated Specification PDF.

v0.3.0

Updated Specification PDF to v0.3.0

Key updates in this version include:

• MCP Server – Updated the MCP Server section with new details and clarifications.
• AI Model Architecture – Revised diagrams and new sections added to reflect the latest architecture.
• Agent Observability – New section added with sub-sections for improved monitoring and insights.
• Issue – New section documenting known issues and tracking improvements.

v0.2.9

Updates

• Updated the Specification PDF to version v0.2.9
• Refreshed and updated the contributors

Documentation

• Introduced new cross-industry Markdown documentation:

  • Oil & Gas - AI-Driven Drilling Intelligence
  • Pharmaceuticals - Sales Engagement Planning
  • Utilities

V0.2.8

Documentation Updates

• Updated the Specification PDF to v0.2.8
• Created new Markdown documentation for:
  • Health Care
  • Oil & Gas
  • Retail
• Agent metadata schema JSON updated including the example JSON's for Fraud detective and Sepsis.

V0.2.6

Agent Identification

• Added ‘Cost Center’ attribute.

Data Sources

• Added ‘Uses PII,’ ‘Uses PHI’ and ‘Uses PCI’ attributes.

Industry Schemas

• Updated Medical Device industry schema (medical-device-manufacturer-schema.json).
• Added Model Risk Management for banking industry schema (banking-model-risk-management.json).

Documentation

• Updated the specification PDF to v0.2.6.
• Updated Markdown document for Medical Device industry support, with links to the corresponding JSON schema.
• Updated Markdown document for Model Risk Management in Banking support, with links to the corresponding JSON schema.
• Updated Markdown document for Enhancing the A2A Protocol, including links to the relevant JSON examples.

v0.2.5

Agent Context Extensions

• Added technical_context to describe relationships between an agent and other technical components (e.g. Agent, LLM Model, Tools, Guardrails).
• Added business_context to capture how agents are used within applications, business processes, and AI use cases.
• Added industry_specific_context to model industry-specific interactions and regulatory environments.
• Added risk_posture to represent assessments, controls, and an overall trust score with rationale.
• Added compliance_posture to represent applicable regulations, policies, and standards, including a compliance score and rationale.

Provider Metadata

• Added provider section to the schema.

Industry Schemas

• Added Medical Device industry schema (medical-device-manufacturer-schema.json).

Documentation

• Updated the specification PDF to v0.2.5.
• Added a Markdown document for Medical Device industry support, with links to the corresponding JSON schema.
• Added a Markdown document for Enhancing the A2A Protocol, including links to the relevant JSON definitions.