AI Agents Inventory and Attributes on The International Foundation for Information Technology (IF4IT)

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

About This Inventory

The AI Agents Inventory is the enterprise’s governed record of its AI agents — autonomous, model-driven software actors that take consequential action, or communicate, on the enterprise’s behalf, each under its own non-human identity. This document is a companion to, and an operational extension of, the Enterprise AI Governance Best Practices document: that document establishes how an enterprise governs its use of AI, while this inventory is the governed record of the individual agents to which that governance must be applied. A single record represents one such agent — part deployed application (versioned, owned, and hosted) and part digital worker (with an identity, a role, access, training, and accountability) — distinct from the AI and Machine Learning Models Inventory, which governs the models an agent is built on, and from the Applications Inventory, which governs the software an agent may be embedded in or act upon.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

The following terms are used throughout this document with specific meanings. Terms defined in the IF4IT Enterprise Inventory Management Best Practices document or in discipline-specific Best Practices documents (APM, TPM, etc.) are not duplicated here — refer to those documents for broader inventory governance and discipline-specific vocabulary.

Term	Abbreviations and Acronyms	Definition
Agentic AI		The paradigm of AI systems that plan and take actions toward goals, as distinct from AI that only predicts or generates content.
AI Agent		An autonomous, model-driven software actor that takes consequential action — operating on systems and data, and/or communicating with stakeholders by any means — on the enterprise’s behalf without a human approving each action, under its own non-human identity.
Attestation		A named, evidence-backed sign-off that an AI agent was properly trained, tested, and authorized before deployment; the defensible record that reasonable care was taken.
Autonomy Level		The degree of independent action an agent is permitted, from advisory (proposes only) through human-approved and supervised to fully autonomous.
Catalog		A discovery and exposure construct built over an Inventory/Registry that surfaces registered Noun Instances and their associated data, content, and services at both the catalog level and the individual instance level. Serves as a knowledge management user interface into the Enterprise Model — letting a person find, learn about, and engage with instances, browse and filter them by their governed attribute values, and traverse their relationships to related Noun Instances.
Catalog of Catalogs		The enterprise-wide federated discovery surface spanning every Noun Type Catalog, exposing registered instances and their associated data, content, and services across the Enterprise Model. The discovery-layer counterpart to the Inventory of Inventories.
Copilot		An AI assistant whose every output a human reviews and approves before it takes effect; it assists rather than acts, and is therefore outside the scope of this inventory.
Evaluation	Eval	Scored testing of an agent’s outputs — for task success, correctness, groundedness, and safety or policy compliance — run continuously rather than once.
Generative AI	GenAI	AI that produces content (text, images, code) in response to prompts; distinct from agentic AI, which acts. An agent is typically built on a generative model but is governed as an actor, not as the model.
Guardrail		A runtime constraint, applied before or after the model, that filters, blocks, or bounds what an agent may do or say.
Human-in-the-Loop	HITL	An arrangement in which a human reviews and approves each output before it reaches a system or stakeholder — the test that distinguishes a copilot from an agent.
MCP Server	MCP	A standardized server that exposes tools, resources, and context to AI agents over an open protocol (currently the Model Context Protocol); reached by an agent through an integration and governed as its own Noun Type.
Model		The AI or machine learning model an agent is built on and driven by, governed in the AI and Machine Learning Models Inventory. The model is not the agent — one model can power many agents.
Multi-Agent System	MAS	A coordinated set of agents that work together; governed through the relationships among the individual agent records, not as a single record.
Non-Human Identity	NHI	The distinct machine identity and credentials — API keys, OAuth grants, service accounts — under which an agent authenticates and acts. An agent has a non-human identity but is more than one.
Platform-Native Registry		A vendor or platform’s internal list of the agents created within that environment; a governed source that seeds the enterprise inventory, never the enterprise system of record.
Prompt Injection		An attack that smuggles adversarial instructions into an agent’s inputs to subvert its intended behavior.
Reasonable Care		The legal standard that an organization take reasonable care to ensure its agents’ representations are accurate and not misleading; the duty an AI Agents Inventory helps an enterprise discharge and document.
Registry		A governed, authoritative collection in which every instance of a Noun Type is formally registered, either manually or through automation. Equivalent to a Noun Inventory — the two terms name the same construct.
Shadow Agent		An AI agent operating without a governed inventory record — unregistered, unowned, and unreviewed; typically the highest-risk agents in an environment.
Stop Authority		The named person or role with the standing right to halt, pause, or roll back an agent in production.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

The AI Agents Inventory governs the enterprise’s AI agents. An AI agent is an autonomous, model-driven software actor that takes consequential action on the enterprise’s behalf — operating on systems and data, and/or producing content that reaches stakeholders by any means, whether a live conversation, a generated email or letter, a document, a notification, or a posted message; electronic or physical; synchronous or asynchronous — in pursuit of a goal, without a human approving each individual action or message, and under its own non-human identity. A single record in this inventory represents one such agent. The defining test is where authority sits: if a human reviews and approves each output before it reaches a system or a stakeholder, the software is a copilot and is governed as a feature of its host application, not here; if the agent’s outputs reach systems or stakeholders without per-output human approval, it is an AI agent and belongs in this inventory.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

The AI Agents Inventory is equivalent to the AI Agents Registry. The two names describe the same governed construct: the authoritative collection in which every AI Agent instance is formally registered. Registration is the act of entering a governed instance into the collection, and it is what makes the collection an authoritative system of record rather than an informal list. Instances are registered either manually, by a practitioner, or through automation that harvests and registers instances from authoritative sources. However registration occurs, a registered instance is a governed record carrying a Semantic Identifier and its governed attribute values.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

The most immediate reason an enterprise must govern its AI agents is accountability. When an AI agent acts or speaks on the enterprise’s behalf, the enterprise owns the consequences. A civil resolution tribunal made this concrete when it held a company liable for a negligent misrepresentation that its customer-facing chatbot made about the company’s own policy, rejecting the argument that the chatbot was a separate entity responsible for its own statements; the standard the tribunal applied was whether the company had taken reasonable care to ensure the agent’s representations were accurate and not misleading. Reasonable care cannot be applied to an agent no one knows exists. An unregistered agent that transacts or communicates on the enterprise’s behalf is therefore uncontrolled legal, regulatory, and reputational exposure — and a governed inventory is the precondition for constraining, overseeing, and defending it.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

The AI Agents Inventory is one of the most connected inventories in the Enterprise Model; the relationships summarized here are treated in full in the closing Relationships to Other Inventories section. Its most defining relationship is to the AI and Machine Learning Models Inventory: every agent is built on and driven by a model, a consumption relationship carried by the agent’s model and model-version attributes. Through Integrations, the inventory cross-references the tools, APIs, and MCP servers an agent acts through; through the Applications Inventory, it cross-references the applications an agent is embedded in or acts upon.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

The AI Agent is one of the recognized Noun Types in the IF4IT Enterprise Inventory Management taxonomy — the master catalog of every enterprise inventory type the framework governs. For the full taxonomy, the governance principles that apply to every inventory, and the Inventory of Inventories that catalogs them, refer to the IF4IT Enterprise Inventory Management Best Practices document.

The AI Agent is among the most connected Noun Types in the taxonomy, and its closest neighbors fall into three bands. Its core relationships are five: it is built on and driven by an AI or machine-learning model governed in the AI and Machine Learning Models Inventory; it is embedded in, exposed by, or acts upon software governed in the Applications Inventory; it acts through the tools, APIs, and connections governed in the Integrations Inventory; it accesses, reads, and writes data governed in the Data and Information Inventory; and it is owned by, accountable to, and attested by people governed in the People and Personnel Inventory.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Section A — Sourcing and Harvesting

Before building the AI Agents Inventory from scratch, assess how much of it can be harvested from systems already operating in the enterprise. AI agents almost always exist — at least partially — in other platforms before a formal inventory is established. The most valuable sources are the platform-native registries maintained by the agent-building and orchestration platforms the enterprise already uses, which list the agents created within each; the identity and access systems that hold the non-human identities and credentials agents authenticate with; the cloud and API gateways that see the calls agents make; and the integration and MCP-server catalogs that reveal what agents connect to. Harvesting from these sources reduces the initial data-entry burden, accelerates time to Crawl completeness, and — critically for agents — surfaces records that manual discovery would miss.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

AI agents raise governance challenges that do not fit neatly into the standard attribute categories — challenges specific to autonomous, model-driven actors, and moving quickly as the field matures. This chapter names the most consequential of them and, for each, states how a governed AI Agents Inventory addresses it. The point is not that the inventory solves these challenges on its own, but that it is the precondition for addressing any of them — which is why this inventory is positioned as the operational extension of the Enterprise AI Governance Best Practices document, the record its agent-level governance is applied against.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Descriptive attributes capture the core identity of each AI agent: how it is identified, named, and described as a distinct governed actor.

Attribute Name

Maturity

Description and Notes

UID

Crawl

Description — A machine-readable, system-generated identifier from any source platform that created or imported the agent record.

Benefit(s) — Supports integration and reconciliation with the platform-native registries and identity systems an agent is harvested from.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Classification attributes categorize each agent along the orthogonal dimensions that determine its governance profile.

Attribute Name

Maturity

Description and Notes

Agent Type

Crawl

Description — The primary purpose/function class of the agent.

Benefit(s) — Drives baseline expectations for risk, oversight, and the controls a class of agent typically needs.

Source — Manual.

Examples — Personal Productivity; Work/Process Automation; Stakeholder Interaction; System/Operations; Data/Analytics; Developer/Engineering; Decision Support/Advisory; Orchestration/Supervisory

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Ownership and stakeholder attributes assign accountability for each agent — including who owns it, who supports it, and who can stop it.

Attribute Name

Maturity

Description and Notes

Business Owner

Crawl

Description — The individual or function accountable for the agent’s purpose, value, and business outcomes.

Benefit(s) — Establishes business accountability and a decision-maker for the agent’s continued operation.

Source — Manual.

Notes — Resolve to a Person Semantic ID as the People and Personnel Inventory matures.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Lifecycle and status attributes govern each agent’s state over time, from proposal through active operation to retirement and archival.

Attribute Name

Maturity

Description and Notes

Lifecycle Status

Crawl

Description — The agent’s current governed lifecycle state.

Benefit(s) — Distinguishes active agents from proposed, deprecated, or retired ones for governance and reporting.

Source — Manual.

Examples — Proposed; Active; Under Review; Deprecated; Retired; Archived

Deployment Date

Walk

Description — The date the agent was first placed into production.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Governance attributes tie each agent to the policies, regulations, and governing bodies that constrain and oversee it.

Attribute Name

Maturity

Description and Notes

Governing Body

Crawl

Description — The committee or function that governs the agent — for example an AI review board or architecture governance body.

Benefit(s) — Establishes where oversight decisions about the agent are made and recorded.

Source — Manual.

Examples — AI Governance Board

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Strategic attributes position each agent within the enterprise’s portfolio and investment intent.

Attribute Name

Maturity

Description and Notes

Strategic Disposition

Walk

Description — The portfolio stance toward the agent.

Benefit(s) — Drives investment, consolidation, and retirement decisions from governed data rather than anecdote.

Source — Manual.

Examples — Invest; Maintain; Consolidate; Retire

Business Value

Walk

Description — The value the agent delivers — qualitative or quantified.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Assessment and health attributes report each agent’s current reliability, evaluated behavior, and known weaknesses.

Attribute Name

Maturity

Description and Notes

Trust Signal

Run

Description — A single, continuously updated indicator of the agent’s current health, reliability, and governance posture.

Benefit(s) — Gives everyone — auditor, risk officer, engineer — one current read on whether the agent can be relied on.

Source — Calculated.

Examples — Green; Amber; Red

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Technical attributes capture how each agent is constructed and configured to reason and act.

Attribute Name

Maturity

Description and Notes

Underlying Model

Crawl

Description — The AI or machine-learning model the agent is built on and driven by.

Benefit(s) — Exposes model dependency and concentration risk; the most defining technical relationship.

Source — Derived.

Notes — Links to the AI and Machine Learning Models Inventory. The model is not the agent.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Operational attributes govern how each agent behaves and is controlled while running in production.

Attribute Name

Maturity

Description and Notes

Business Criticality

Walk

Description — How critical the agent is to business operations.

Benefit(s) — Prioritizes oversight, monitoring, and incident response.

Source — Manual.

Examples — Low; Medium; High; Mission-Critical

Runtime Observability

Walk

Description — Whether the agent’s decisions, tool calls, and outputs are traced and monitored in production.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Security attributes govern each agent’s identity, credentials, access, and protection against attack.

Attribute Name

Maturity

Description and Notes

Non-Human Identity

Crawl

Description — The dedicated machine identity the agent authenticates and acts under, scoped to least privilege.

Benefit(s) — Makes the agent individually attributable and revocable, and its access reviewable — no shared or accumulating credentials.

Source — Manual.

Examples — Dedicated service identity

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Data and information attributes record the data each agent accesses, retains, and produces.

Attribute Name

Maturity

Description and Notes

Data Accessed

[Multi-Value]

Crawl

Description — The data stores and datasets the agent can access, read, or write.

Benefit(s) — Reveals the agent’s data reach — essential for security, privacy, and impact analysis.

Source — Derived.

Examples — Orders datastore; Customer profile store

Notes — Links to the Data and Information and Data Stores inventories.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Geographic and jurisdictional attributes record where each agent operates and whose rules therefore apply.

Attribute Name

Maturity

Description and Notes

Hosting Region(s)

[Multi-Value]

Walk

Description — The geographic regions where the agent runs.

Benefit(s) — Establishes the jurisdictions governing the agent’s operation.

Source — Derived.

Examples — US-East; EU-West

Serving Geographies

[Multi-Value]

Walk

Description — The regions whose stakeholders the agent interacts with.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

No standard temporal attributes are identified for the AI Agents Inventory; the dates that matter are governed within their owning categories.

The IF4IT has not yet identified standard attributes in this category for the AI Agents Inventory. This does not mean no such attributes exist — organizations building this inventory are encouraged to identify and define attributes in this category that are relevant to their specific context and governance needs.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Contractual and legal attributes record the terms of use and liability allocation governing each agent.

Attribute Name

Maturity

Description and Notes

Model and Platform License

[Multi-Value]

Walk

Description — The licenses and terms governing use of the agent’s underlying model and platform.

Benefit(s) — Surfaces usage restrictions and obligations that constrain how the agent may be deployed.

Source — Derived.

Examples — Vendor commercial license; platform EULA

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Vendor and supplier attributes identify the model provider, platform provider, and builder behind each agent.

Attribute Name

Maturity

Description and Notes

Model Provider

Crawl

Description — The vendor that provides the agent’s underlying model.

Benefit(s) — Exposes model-vendor concentration and lock-in across the agent estate.

Source — Derived.

Notes — Links to the Vendors Inventory.

Platform Provider

Crawl

Description — The vendor that provides the platform or framework the agent is built and run on.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

IT environment attributes record the operating environments each agent is deployed to.

Every attribute in this category is Calculated or Derived from records in other inventories. No manual data entry is required for any attribute in this category. Do not create data entry fields for any attribute listed here.

Attribute Name

Maturity

Description and Notes

Deployment Environments

[Multi-Value]

Walk

Description — The IT operating environments the agent is deployed to and runs in.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Skills and competencies attributes record the human expertise each agent depends on.

Attribute Name

Maturity

Description and Notes

Skills Required to Build and Maintain

[Multi-Value]

Run

Description — The competencies needed to build, operate, evaluate, and govern the agent.

Benefit(s) — Reveals the human capacity an agent depends on and supports workforce planning.

Source — Manual.

Examples — Prompt engineering; MLOps; AI risk review

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Provenance and audit attributes record how each agent’s record was created and where the trace of its behavior is kept.

Attribute Name

Maturity

Description and Notes

Creation Timestamp

Walk

Description — When the agent’s record was created — also its registration date in the inventory.

Benefit(s) — Anchors audit history and supports the Catalog’s browse-by-registration-date.

Source — Derived.

Execution-Trace and Event-Logging Reference

Walk

Description — Where the tamper-evident log of the agent’s decisions, tool calls, and actions is kept.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Risk attributes classify and rate the specific ways each agent can cause harm.

Attribute Name

Maturity

Description and Notes

Regulatory Risk Tier

Walk

Description — The agent’s risk classification under applicable AI regulation.

Benefit(s) — Determines the mandatory controls, oversight, and record-keeping the agent requires.

Source — Manual.

Examples — Prohibited; High-Risk; Limited-Risk; Minimal-Risk

Notes — Aligned to frameworks such as the EU AI Act; links to Regulations.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Compliance and regulatory attributes record whether each agent meets the obligations that apply to it.

Attribute Name

Maturity

Description and Notes

Applicable Regulatory Frameworks

[Multi-Value]

Walk

Description — The regulatory and standards frameworks the agent must comply with.

Benefit(s) — Establishes the agent’s compliance scope.

Source — Derived.

Examples — EU AI Act; ISO/IEC 42001; NIST AI RMF

Notes — Links to Regulations and Regulatory Obligations.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Financial attributes govern the cost of running each agent and who is accountable for it.

Attribute Name

Maturity

Description and Notes

Model and Usage Cost

Walk

Description — The model, token, and tool-invocation cost the agent incurs, typically measured per period.

Benefit(s) — Makes the variable cost of an autonomous agent visible and manageable.

Source — Calculated.

Total Cost of Ownership

Run

Description — The fully loaded cost of the agent — usage, platform, build, and oversight.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Assurance and attestation attributes record the evidence-backed sign-off that each agent was authorized as fit for its purpose and autonomy — the enterprise’s defensibility record.

Attribute Name

Maturity

Description and Notes

Fitness Attestation Status

Crawl

Description — Whether a named authority has attested the agent is fit for its purpose and permitted autonomy.

Benefit(s) — Records that the agent was formally judged safe to operate — the core of the defensibility record.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Autonomy and behavioral controls attributes define and enforce the limits on each agent’s autonomous behavior.

Attribute Name

Maturity

Description and Notes

Autonomy-Level Enforcement

Crawl

Description — How the agent’s permitted autonomy level (from Classification) is technically enforced.

Benefit(s) — Ensures the declared autonomy level is a real boundary, not just a label.

Source — Manual.

Examples — Approval gate; scoped permissions; action allowlist

Notes — The autonomy level itself is set in Classification.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

Relationship attributes summarize each agent’s connective footprint across the Enterprise Model as Derived and Calculated values.

Attribute Name

Maturity

Description and Notes

Total Connected Integrations Count

Run

Description — The number of integrations the agent acts through.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

The attribute categories presented above represent the IF4IT’s suggested baseline for governing the AI Agents Inventory. They are designed to cover the attribute dimensions that matter to almost every enterprise — but they are not exhaustive, and they are not mandatory. Every enterprise operates within a unique context, governed by its own regulatory environment, organizational structure, industry dynamics, and strategic priorities.

This document extends the standard baseline with two custom categories — Assurance and Attestation, and Autonomy and Behavioral Controls — because autonomous, model-driven agents require governance dimensions that a general attribute baseline does not provide: a defensible record that an agent was attested fit to operate, and an explicit definition of the limits on its autonomous behavior. Both follow the same structural pattern as every other category and are placed before the Relationship Attributes category, which always remains last.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

The relationship is one of consumption: every AI agent is built on and driven by a model governed in the AI and Machine Learning Models Inventory, which is authoritative for the model’s identity, capabilities, and lifecycle. The AI Agents Inventory references the model rather than describing it — the model is a distinct Noun Type, and one model can drive many agents. The AI and Machine Learning Models Inventory is not yet published; until it is, refer to the IF4IT Enterprise Inventory Management Best Practices document for its current definition.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

The relationship is a cross-reference between distinct Noun Types: an AI agent may be embedded in an application, exposed through one, or act upon one, while the Applications Inventory remains authoritative for the applications themselves. Neither seeds the other; each references the other where they meet.

The connecting attributes are the agent’s Tools and Functions Available and the application references it acts upon or is embedded in, on the AI Agents side, matched to the application records in the Applications Inventory.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

The relationship is a cross-reference: an AI agent acts through the tools, APIs, and connections governed in the Integrations Inventory, which is authoritative for those integrations. An agent’s connections to external tools and MCP servers are realized as integrations and referenced here.

The connecting attributes are the agent’s Tools and Functions Available and Communication Protocols in the Technical category, and its Total Connected Integrations Count in the Relationship category, which quantifies the action surface.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

The relationship is one of consumption: an AI agent accesses, reads, and writes data governed in the Data and Information Inventory, which is authoritative for those data assets. The AI Agents Inventory records what each agent touches; the data inventory records what the data is.

The connecting attributes are the agent’s Data Accessed and Data Classification Handled in the Data and Information category, and its Total Data Stores Accessed in the Relationship category. Data classification is resolved through the Data Sensitivity Types Inventory.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

The relationship is a cross-reference to accountable people: an AI agent is owned by, accountable to, stoppable by, and attested by individuals governed in the People and Personnel Inventory, which is authoritative for those people. The People and Personnel Inventory is not yet published; until it is, refer to the IF4IT Enterprise Inventory Management Best Practices document for its current definition.

The connecting attributes are the agent’s Business Owner, Technical Owner, Accountable Individual, and Stop Authority in the Ownership and Stakeholder category, and its Attesting Authority in the Assurance and Attestation category — each populated as a name at Crawl and resolved to a Person Semantic Identifier as the People and Personnel Inventory matures.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

The relationship is one of dependency: an AI agent is built and run on frameworks and runtimes governed in the Software Technologies Inventory, which is authoritative for those technologies. This is distinct from the agent’s model dependency, which the AI and Machine Learning Models relationship covers. The Software Technologies Inventory is not yet published; until it is, refer to the IF4IT Enterprise Inventory Management Best Practices document for its current definition.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

The relationship is a cross-reference: an AI agent automates or delivers one or more business capabilities governed in the Capabilities Inventory, which is authoritative for those capabilities. The agent references the capabilities it serves.

The connecting attribute is the set of capability references the agent automates, matched to the capability records in the Capabilities Inventory.

Maintaining this relationship reveals where the enterprise’s capabilities are becoming agent-automated — supporting capability-level risk analysis, redundancy detection, and investment decisions. Without it, the shift of capabilities to autonomous execution goes untracked.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

The relationship is a cross-reference: an AI agent is deployed to and runs in one or more IT operating environments governed in the Environments Inventory, which is authoritative for those environments. The Environments Inventory is not yet published; until it is, refer to the IF4IT Enterprise Inventory Management Best Practices document for its current definition.

The connecting attributes are the agent’s Deployment Environments and Environment Type in the IT Environment category, which resolve to the environment records governed in the Environments Inventory.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

The relationship is a cross-reference across three vendor roles: an AI agent’s model provider, platform provider, and builder are each governed in the Vendors Inventory, which is authoritative for vendor facts. The builder role may resolve to the enterprise itself for internally built agents.

The connecting attributes are the agent’s Model Provider, Platform Provider, and Builder in the Vendor and Supplier category, each resolving to a vendor record (or to Self).

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

The relationship is one of constraint: an AI agent is bound by laws, regulations, and directives governed in the Regulations Inventory, which is authoritative for them and covers laws, regulations, and directives alike. It is also the gateway to the Regulatory Obligations and Regulatory Agencies relationships. The Regulations Inventory is not yet published; until it is, refer to the IF4IT Enterprise Inventory Management Best Practices document for its current definition.

The connecting attributes are the agent’s Related Regulations in the Governance category, its Applicable Regulatory Frameworks in the Compliance and Regulatory category, and its Regulatory Risk Tier in the Risk category.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

The MCP Servers Inventory governs, as first-class assets, the standardized servers an agent connects to over the Model Context Protocol. An agent’s connections to MCP servers are realized as integrations and counted in its Number of MCP Servers Connected attribute; the MCP Servers Inventory is a new Noun Type that governs those servers themselves — their identity, ownership, and the tools they expose.

The Regulatory Obligations and Regulatory Agencies inventories are reached through the Regulations relationship: obligations are the specific duties a regulation imposes on an agent, and agencies are the bodies that enforce them. Together they let an agent’s Regulatory Oversight Status and per-obligation compliance be governed precisely.

AI Agents Inventory and Attributes

Fri, 17 Jul 2026 00:00:00 +0000

The AI Agents Inventory and Attributes document is one of many Noun Type inventories in the IF4IT Enterprise Inventory Management taxonomy. Each inventory in the taxonomy governs a distinct class of enterprise asset, relationship, or organizational construct — and each contributes its records and relationships to the Enterprise Model that connects them all. The full catalog of recognized Noun Types, the principles that govern inventory design and federation, the Inventory of Inventories, and guidance for building, connecting, and maturing a complete enterprise inventory program are documented in the IF4IT Enterprise Inventory Management Best Practices document. Practitioners who find value in the AI Agents Inventory are encouraged to explore the broader taxonomy — each additional inventory that is built and governed strengthens the Enterprise Model and expands the range of questions that AI-assisted enterprise analysis can answer. The IF4IT welcomes contributions from practitioners who develop attribute categories, attribute definitions, or governance approaches not covered in this document — your real-world experience is how this framework matures.