# Command centers for data stewards and context engineers

**What Google Knowledge Studio signals. Business context for agents is becoming an operated system — with owners, review gates, version control and a throttle a human holds.**

Checked: 23 October 2026 · auxfirst
Canonical: https://auxfirst.com/news/google-knowledge-studio-context-command-center.html

> **The short version**
>
> On 23 October 2026 Google Cloud launched two previews inside Knowledge Catalog: *self-learning context enrichment*, which watches how agents and users actually interact with data and proposes fixes to business definitions, and *Knowledge Studio*, a console where data stewards and context engineers explore the data estate, navigate a visual context graph and approve or reject proposed changes before anything reaches an agent. Google calls Knowledge Studio a “command center” for those two roles. That phrase is the real news. Context for AI agents is becoming an operated system with owners, review gates and version control — not a prompt someone wrote once.

Every agent answers from the definitions it was last given. Ask one what net revenue was in EMEA last quarter and the answer depends on whether “net revenue” still means what the finance playbook said in March, which tables hold the bookings, and how they join. That knowledge lives, in most companies, in a Google Doc, a GitHub repo, a Confluence page and three people’s heads. The moment finance changes how partner credits are treated, every agent with a hard-coded formula starts answering wrong, silently and at once.

Google’s announcement names that failure mode and ships a loop against it. [1] This piece explains what was shipped, what a context command center actually has to do, why two roles that used to work in two tools now share one screen, and what the pattern means for brand, marketing and GTM teams whose vocabulary — “qualified lead”, “active customer”, “campaign”, “SKU” — is the context agents reason over.

## 01 · What Google announced on 23 October 2026

The facts below come from Google Cloud’s announcement, dated 23 October 2026 and authored by Firat Elbey, Group Product Manager for Data Analytics, and Vasiya Krishnan, Product Lead. [1]

- **Self-learning context enrichment (preview)** analyzes agent and user interaction traces — from Gemini agents, Conversational Analytics in BigQuery or custom agents, with PII redacted — to find where agents struggled: ambiguous column lookups, ungrounded metrics, missing definitions, users rephrasing the same question. It then finds the authoritative source and stages a fix.
- **Knowledge Studio (preview)** is the redesigned Knowledge Catalog experience for context engineers and data stewards: a visual map of the cross-cloud data estate, conversational exploration of the context graph, a review queue for proposed enrichments with source citations and a confidence score, and monitoring of enrichment runs including token usage and run duration.
- **Context sets** are governed subsets of the enterprise context graph — definitions, metrics, join logic, golden queries — scoped to a domain, team, agent or user group and controlled with IAM. The same catalog can hold a Sales definition of “net revenue” and a Finance definition without conflict.
- **Human control is configurable.** Auto-publishing can be switched on or off. Changes above a steward-set confidence threshold — Google’s example is 85% — are approved automatically; anything below it, and sensitive changes such as deletions, become a Change Request a steward approves, edits or rejects. Every change is logged with provenance and can be reverted.
- **Claimed impact.** Google states that Knowledge Catalog makes agents nearly 4× more accurate, with up to 33% lower latency and up to 75% fewer tokens, “based on internal benchmarks”. No method or external replication was published with the claim.
- **Availability.** The features “will be rolled out to customers in the next few weeks”, with a sign-up form for early access. Pricing was not stated.

Two pieces of lineage matter for anyone searching for this product. Knowledge Catalog is the name Dataplex Universal Catalog has carried since 10 April 2026; the API, client library, CLI and IAM names did not change. [3] And the context sets inside it build on the Open Knowledge Format, which Google published as an open specification on 12 June 2026. [2]

## 02 · What a context command center is

The market does not have a settled definition yet, so here is the one we use at auxfirst.

> **Definition.** *A context command center is the operating console where the people accountable for an organization’s business definitions can see what agents currently know, see where that knowledge is failing in live use, review machine-proposed corrections against their sources, decide what gets published to which agents, and roll back anything that goes wrong — with every decision logged.*

The word *command* is doing real work. A catalog is an inventory: it tells you what exists. A command center is an operations room: it tells you what is going wrong right now, what the system proposes to do about it, and gives a human the throttle. The shift Google describes — from “a repository of static inventory” to “an interactive command center”, in its own words [1] — is the shift infrastructure made a decade ago, when monitoring dashboards turned into incident consoles with runbooks and on-call rotations.

*Figure 1 · Five capabilities separate a command center from a catalog with a nicer interface.*

1. **Coverage view** A map of what context exists, what is stale and what is missing — across warehouses, operational databases, semantic models and documentation, not one cloud.
2. **Failure signal from production** Evidence pulled from real agent interactions showing where agents gave wrong or ungrounded answers. Without it, stewards document by guesswork.
3. **Proposal queue with provenance** Machine-drafted changes that arrive with the source they came from and a confidence score, so the reviewer judges evidence instead of rewriting from scratch.
4. **Policy-driven autonomy** A threshold a human sets that decides which changes publish automatically and which wait — plus hard gates for destructive actions, whatever the score.
5. **Versioned publish and rollback** Software-style version control on business definitions, scoped to audiences, with reversal as a first-class action.
6. **The test** A product with one or two of these is a catalog. Knowledge Studio ships all five in preview, which is why it is worth reading as a category signal rather than a feature update.

auxfirst’s five-capability test. Each maps onto something in Google’s announcement: the visual estate map, interaction-trace analysis, citations and confidence scores on every proposal, the configurable threshold with deletions always gated, and change requests with software-style version control. [1]

## 03 · Why data stewards and context engineers now share one console

Until recently these were two jobs in two tools.

**Data stewards** own meaning: what a metric means, which table is the system of record, which join is correct, which definition is deprecated. Their tools were catalogs, glossaries and governance workflows built for humans reading documentation.

**Context engineers** own what an agent is given to reason with: system prompts, retrieval configuration, tool descriptions, schema snippets, golden queries, and the instruction files — AGENTS.md, CLAUDE.md, [brand.md](https://auxfirst.com/news/brand-md-context-file.html) and their relatives — that shape agent behavior. Their tools were code repositories, prompt editors and evaluation harnesses.

Google’s announcement names the breaking point: context engineers writing prompts and mappings agent by agent, data stewards keeping definitions current across thousands of assets by hand, and neither approach surviving the move from a handful of agents to thousands. [1]

*Figure 2 · Two roles, one object.* The data steward (owns meaning; catalogs, glossaries, governance workflows built for people reading docs) and the context engineer (owns what the agent gets; prompts, retrieval configuration, tool descriptions, evaluation sets built for code and harnesses) both converge on one object: the governed definition, which is also the agent’s context. The steward’s approval is what changes the agent’s behavior; the engineer’s evaluation set is what tests the steward’s definition. A single console forces both roles onto one object: the governed definition that is *also* the agent’s context. That coupling is the design decision, and it is the right one. In our practice we call the place where accountable humans sign what an agent will treat as true the Authorship Layer.

## 04 · How self-learning enrichment works, step by step

Stripped of product language, the loop Google describes has six stages. [1]

*Figure 3 · The enrichment loop, with the one stage a human holds.* 01 Observe: agent and user traces from Gemini agents, BigQuery Conversational Analytics or custom agents, PII redacted. 02 Detect: ambiguous lookups, ungrounded metrics, missing definitions, rephrased questions. 03 Source: Workspace, GitHub, Confluence, SharePoint, Cloud Storage and structured sources. 04 Propose: a drafted change with citations and a confidence score. 05 Gate: above the threshold, publish automatically if auto-publish is on; below it, or any deletion, a Change Request a steward approves, edits or rejects. 06 Publish, log, revert into a governed context set, with full provenance and rollback. Agents read the set and their traces start the next cycle; an organization can also supply its own evaluation set to test proposals end to end. The loop as Google describes it, with product names removed. [1] On every run the system also detects what changed at the source, retires obsolete definitions and flags conflicts. Stages one to four and six run at machine speed; stage five is where the organization’s review capacity sets the real pace.

1. **Observe.** Interaction traces from Gemini agents, BigQuery Conversational Analytics or custom agents are analyzed, with PII redacted.
2. **Detect.** The system flags patterns that indicate missing or outdated context: ambiguous column lookups, metrics with no grounding, repeated rephrasing of the same question.
3. **Source.** It searches connected knowledge — Google Workspace, GitHub, Confluence, SharePoint, Cloud Storage, and first- and third-party structured sources — for the authoritative definition, and links unstructured sources such as playbooks to structured metadata such as tables and schemas.
4. **Propose.** It drafts business definitions, table and column descriptions, join relationships, golden queries and task-specific agent instructions, each with citations and a confidence score.
5. **Gate.** Proposals above the configured threshold publish automatically if auto-publish is on; everything else, and all deletions, becomes a Change Request for a steward. An organization can supply its own evaluation set to test proposals end to end.
6. **Publish, log, revert.** Approved changes land in a governed context set. Every change carries full provenance and can be rolled back.

## 05 · The gate, read on the Action Heat Ladder

Readers of our work on [the Action Heat Ladder](https://auxfirst.com/action-heat-ladder.html) will recognize the structure of stage five. The confidence threshold is a heat setting: it decides how much autonomy the enrichment agent gets per class of change. Deletions are pinned to the hot end regardless of confidence — a hard gate, not a score. Google got that ordering right, and it is the ordering most teams get wrong when they first wire a model’s confidence into a workflow.

*Figure 4 · One threshold, three classes of change — the consequence sets the row.*

|  | Above the threshold — Google’s example: 85% | Below the threshold — any confidence | Regardless of confidence — sensitive changes |
|---|---|---|---|
| Descriptive — a column description, a golden query, a link to a playbook | **Publish, log it.** Cheap to reverse, visible in the diff. | Change Request. A steward approves, edits or rejects. | — |
| Definitional — what “net revenue” means, which join is correct, a retired definition | Publish to the scoped set, log it, sample the reviews. | Change Request, tested against the evaluation set first. | — |
| Destructive — a deletion, a conflict between two sources | Change Request. The score does not apply. | Change Request. | **Always a person.** This is Google’s hard gate for deletions. |

auxfirst’s reading of the announced controls, not Google’s product documentation. Google describes one threshold plus a hard gate for deletions; the row split is ours, and it is the split we would ask a steward to configure. A high score on a destructive change is still a Change Request. [1]

The second thing readers will recognize is the Change Request queue. It is the *verification throughput* constraint we described in [the Publisher Agent Map](https://auxfirst.com/ad-industry/publisher-agent-map.html): the number of outputs a qualified person can genuinely check in a day. Enrichment proposes faster than stewards can review, so the real capacity of the loop is the review bench, not the model. A team that raises its threshold to clear the queue is not getting more automation; it is getting less review.

> **The failure mode to design out first**
>
> A threshold is a parameter of the mandate, not of the product. Who may approve a definition of “active customer”, in writing, per class of change, is the [AI agent operating model](https://auxfirst.com/agent-owners-manual.html) for the enrichment agent itself. Set the number without naming the owner and the queue becomes an inbox nobody is accountable for clearing — which is the state most glossaries are in today, with better tooling.

## 06 · Context sets: scoping context per audience

A context set is the unit of publication. Google positions it as a curated, governed subset of the enterprise context graph for a specific domain, team, agent or group of end users, with IAM controlling who can read, use or edit it. [1] A sales agent bound to the Sales set and a finance analyst’s assistant bound to the Finance set can each hold a legitimate, different definition of the same term. Agents are not locked to one set; through search in Knowledge Catalog an agent can locate the right set for a task.

*Figure 5 · One graph, several sets, the same word meaning two things.* The enterprise context graph (definitions, metrics, join logic, golden queries, across clouds) is split into governed subsets. Sales context set: net revenue = bookings − discounts, bound to the sales agent, IAM: sales ops reads, steward edits. Finance context set: net revenue = recognized − partner credits, bound to the finance analyst’s assistant, IAM: finance reads, controller edits. A set tells an agent how to interpret data; access to the data itself still follows existing permissions. Any set can be exported as an OKF bundle, and an OKF bundle can be imported as a set. The two definitions of “net revenue” are illustrative; the mechanism — one catalog, per-domain sets, IAM on each, agents that can search for the right set — is as announced. [1]

Two properties matter for governance teams. **Access to the underlying data still follows existing permissions**: a context set tells an agent how to interpret data, it does not grant access to it. And **portability**: context sets build on the Open Knowledge Format, so an OKF bundle can be imported as a set and a set exported back out. Google states this explicitly as the way to keep context from being locked to its catalog. [1]

## 07 · The Open Knowledge Format: the portable substrate underneath

The Open Knowledge Format is an open specification Google Cloud published on 12 June 2026, authored by Sam McVeety and Amir Hormati. [2] It represents knowledge as a directory of Markdown files with YAML front matter. Each file is one *concept* — a table, a dataset, a metric, a playbook, a runbook, an API — and the file path is the concept’s identity. Concepts link to each other with ordinary Markdown links, which turns the directory into a graph. The only required front-matter field is `type`; title, description, resource and tags are optional. Optional `index.md` files support progressive disclosure as agents navigate; optional `log.md` files hold change history. [4]

*Figure 6 · Anatomy of an OKF bundle.*

```
# a bundle is a directory; the unit of distribution
acme-retail/
├── index.md              # optional: directory listing for progressive disclosure
├── log.md                # optional: chronological history of updates
├── metrics/
│   ├── index.md
│   └── net-revenue.md    # one concept = one file; the path is its identity
├── tables/
│   └── orders.md
└── playbooks/
    └── partner-credits.md

# metrics/net-revenue.md
---
type: Metric                      # the only required field
title: Net revenue
description: Bookings net of discounts and partner credits
resource: bq://acme.finance.net_revenue_v3
tags: [finance, revenue]
---
Net revenue is computed from [orders](../tables/orders.md)
per the [partner credits playbook](../playbooks/partner-credits.md).
```

Structure and field names from the OKF specification; the content is illustrative. Version 0.1 was published on 12 June 2026; by October the specification had reached version 0.2 and moved to its own repository, with a reference enrichment agent for BigQuery, a single-file visualizer and four sample bundles. [4]

Google frames OKF as a formalization of the “LLM wiki” pattern — Obsidian vaults wired to coding agents, the AGENTS.md and CLAUDE.md family of convention files, repositories full of index.md and log.md artifacts, metadata-as-code inside data teams — so that wikis written by different producers can be consumed by different agents without translation. [2]

Why this matters to the command-center argument: the console is Google’s, but the substrate is files. If a steward’s approved definitions can leave as a directory of Markdown that any other tool can read, the command center is a control plane over knowledge the organization still owns. That is the condition under which it is safe to let a vendor’s console become the place where business truth is edited. We have argued the same point from the other direction in [Markdown for AI](https://auxfirst.com/news/markdown-for-ai.html) and in the template for [AGENTS.md for teams that don’t write code](https://auxfirst.com/news/agents-md-for-non-engineering-teams.html): business context belongs in plain, versioned, human-readable files, and the tooling around them should be swappable.

## 08 · What this means for brand, marketing and GTM teams

The example Google uses is a finance question. Swap in your own vocabulary and the stakes are identical.

*Figure 7 · Five things to expect, and what to do about each.*

1. **Your definitions are the context** “Qualified lead”, “active brand”, “share of shelf”, “campaign”, “household” — each has a house definition in a deck, a wiki page and three heads. A command center is where it becomes governed, scoped and current.
2. **Stewardship is moving out of IT** The person who knows what “active customer” means in your market sits in marketing operations or RevOps. The review queue is designed for that person. Expect to be asked to staff it.
3. **Context sets map to the org chart** Brand, retail media, CRM, trade marketing and finance can each hold their own definitions of overlapping terms without a global glossary nobody follows. The cost: someone owns each set.
4. **The review bench is the bottleneck** Enrichment proposes at machine speed; approvals happen at human speed. Set thresholds per class of change and reserve human time for the hot end, or the queue becomes an inbox.
5. **Portability is a procurement question** Ask any vendor selling a “context layer” whether your definitions export as plain files. If the answer is no, you are buying a catalog, not a command center.
6. **Start with the file you already own** A brand.md or an AGENTS.md written this quarter is a context set in waiting. Run it through the [Context File Checker](https://auxfirst.com/tools/context-file-checker.html) and it is ready to import when the console arrives.

Where this sits in a wider program: the processes and ownership layers of [Five Steps to Becoming an Agentic Organization](https://auxfirst.com/5steps.html), and the part of [agent-first design](https://auxfirst.com/agent-first-design.html) that decides what an agent is allowed to treat as true.

## 09 · What is known, what is claimed, what is unknown

| Claim | Source | Standing |
|---|---|---|
| Self-learning enrichment and Knowledge Studio exist, in preview, announced 23 October 2026 | Google Cloud announcement [1] | Announced |
| Configurable threshold, hard gate for deletions, provenance and rollback on every change | Google Cloud announcement [1] | Announced |
| Context sets import from and export to OKF | Google Cloud announcement [1] | Announced |
| Knowledge Catalog is the renamed Dataplex Universal Catalog; API, CLI and IAM names unchanged | Google Cloud documentation, 10 April 2026 [3] | Documented |
| OKF: Markdown with YAML front matter, only `type` required, spec at v0.2 in its own repository | OKF specification [2][4] | Published |
| Nearly 4× accuracy, up to 33% lower latency, up to 75% fewer tokens | Google, “based on internal benchmarks” [1] | Vendor claim, no method |
| Rollout “in the next few weeks” | Google Cloud announcement [1] | Preview, may change |
| Pricing for the two features | — | Not stated |
| OKF adoption outside Google’s own tooling | — | Not documented in the sources read |
| The full connector list for structured sources | Product documentation, not the announcement | Not in the sources read |

## The AUX takeaway

The interesting word in Google’s announcement is not *self-learning*. It is *command*. A system that proposes changes to what your agents believe, at machine speed, needs a room with a human in it, a throttle that human controls, and a log that shows what was changed and by whom. Google has built the room. What it cannot build for you is the roster: who sits in it, which classes of change they own, and what happens when the queue is longer than their day.

> *A catalog tells you what exists. A command center tells you what is going wrong, what the system proposes, and who gets to say yes.*

Business context was always an operated system. It was just operated badly, in decks and heads. The announcement makes the operations visible, which is the first step to making them accountable — and that, not the 4×, is the part worth staffing for.

## Terms used here

**Context command center** — auxfirst's term for the operating console where the people accountable for an organization's business definitions can see what agents currently know, see where that knowledge fails in live use, review machine-proposed corrections against their sources, decide what is published to which agents, and roll back anything that goes wrong, with every decision logged.

**Self-learning context enrichment** — A Knowledge Catalog capability, in preview from 23 October 2026, that analyzes agent and user interaction traces to find missing or outdated business context, finds the authoritative definition in connected documents, repositories and tables, and proposes a fix with citations and a confidence score.

**Context set** — A curated, governed subset of the enterprise context graph, holding the definitions, metrics, join logic and golden queries for one domain, team, agent or user group, with IAM deciding who can read, use or edit it. Context sets build on the Open Knowledge Format and can be imported from and exported to it.

**Context engineer** — The role that decides what an AI agent is given to reason with: definitions, schema context, tool descriptions, golden queries and instruction files, and that tests whether that context produces correct behavior.

**Data steward** — The role that owns meaning: what a metric means, which table is the system of record, which join is correct, which definition is deprecated. In a command center, the steward's approval is what changes the agent's behavior.

**Open Knowledge Format (OKF)** — An open specification from Google Cloud, first published 12 June 2026 and at version 0.2 by October, that represents business knowledge as a directory of Markdown files with YAML front matter. Each file is one concept, links between files form a graph, and only a type field is required.

**Change Request** — In Knowledge Catalog, a proposed context change that did not clear the steward-set confidence threshold, or that is sensitive such as a deletion, and therefore waits for a steward to approve, edit or reject it.

**Verification throughput** — auxfirst's term for the number of agent outputs a qualified person can genuinely check in a day. In an enrichment loop it is the real capacity of the system, because proposals arrive at machine speed and approvals happen at human speed.

## Questions people are asking this week

### What is Google Knowledge Studio?

Knowledge Studio is the redesigned, preview interface of Google Cloud's Knowledge Catalog, announced on 23 October 2026. It gives data stewards and context engineers a visual map of the cross-cloud data estate, conversational exploration of the enterprise context graph, a review queue for machine-proposed context changes with source citations and confidence scores, and monitoring of enrichment runs including token usage and run duration.

### What is self-learning context enrichment?

A preview capability in Knowledge Catalog that analyzes agent and user interaction traces, with PII redacted, to find missing or outdated business context, locates the authoritative definition in connected documents, repositories and tables, and proposes a fix with citations and a confidence score. Depending on a steward-set threshold the fix is published automatically or becomes a Change Request for a person to approve, edit or reject.

### What is the difference between a data catalog and a context command center?

A catalog records what data assets exist and what they mean. A command center adds live failure signals from agent use, a proposal queue with provenance, a human-set autonomy threshold with hard gates for destructive changes, scoped publication to specific agents, and versioned rollback. Knowledge Studio is Google's move from the first to the second.

### What does a context engineer do?

A context engineer decides what an AI agent is given to reason with: definitions, schema context, tool descriptions, golden queries and instruction files such as AGENTS.md, and tests whether that context produces correct behavior. In a command-center model the context engineer's evaluation sets test the data steward's approved definitions.

### What is the Open Knowledge Format?

An open specification from Google Cloud, first published on 12 June 2026, that represents business knowledge as a directory of Markdown files with YAML front matter. Each file is one concept, links between files form a graph, and only a type field is required. The specification is at version 0.2 and lives in its own GitHub repository with a reference agent, a visualizer and four sample bundles. Knowledge Catalog can import OKF bundles as context sets and export context sets as OKF.

### Is Knowledge Catalog the same as Dataplex?

Knowledge Catalog is the name Dataplex Universal Catalog has carried since 10 April 2026. According to Google's documentation the API, client library, CLI and IAM names did not change.

### Is Knowledge Studio available now, and what does it cost?

Both features were announced as previews rolling out to customers in the next few weeks from 23 October 2026, with a sign-up form for early access. The announcement does not state pricing for self-learning enrichment or Knowledge Studio, and the performance figures it gives, nearly 4 times the accuracy with up to 33% lower latency and 75% fewer tokens, are Google's internal benchmarks with no published method.

Method · Every source was fetched and read on 23 October 2026. Product facts come from Google’s two announcements and its documentation; nothing is claimed from the demo video, which is cited only as the demo Google itself links. Vendor performance figures are labeled as vendor figures wherever they appear. The five-capability test, the row split in Figure 4 and the two definitions of net revenue in Figure 5 are auxfirst’s, not Google’s. Product details reflect the announcement and will be revised when the features reach general availability.

## Sources

1. [Introducing new self-learning context enrichment in Knowledge Catalog](https://cloud.google.com/blog/products/data-analytics/knowledge-catalog-self-learning-context-enrichment-for-ai-agents) — Firat Elbey and Vasiya Krishnan, Google Cloud Blog, 23 October 2026.
2. [Introducing the Open Knowledge Format](https://cloud.google.com/blog/products/data-analytics/how-the-open-knowledge-format-can-improve-data-sharing) — Sam McVeety and Amir Hormati, Google Cloud Blog, 12 June 2026.
3. [Knowledge Catalog overview](https://docs.cloud.google.com/knowledge-catalog/docs/introduction) — Google Cloud documentation (rename notice dated 10 April 2026), read 23 October 2026.
4. [Open Knowledge Format — specification v0.2, reference agent and sample bundles](https://github.com/GoogleCloudPlatform/open-knowledge-format) — GitHub, read 23 October 2026.
5. [How Knowledge Catalog’s self-learning enrichment grounds agents](https://www.youtube.com/watch?v=qeWzgLGSkls) — Google Cloud demo video, linked from the announcement.

Related on auxfirst: [the Action Heat Ladder](https://auxfirst.com/action-heat-ladder.html) · [the Publisher Agent Map](https://auxfirst.com/ad-industry/publisher-agent-map.html) · [Markdown for AI](https://auxfirst.com/news/markdown-for-ai.html) · [AGENTS.md for teams that don’t write code](https://auxfirst.com/news/agents-md-for-non-engineering-teams.html) · [brand.md](https://auxfirst.com/news/brand-md-context-file.html) · [What is an MD file](https://auxfirst.com/news/md-file-markdown-guide.html) · [Context File Checker](https://auxfirst.com/tools/context-file-checker.html) · [AI agent operating model](https://auxfirst.com/agent-owners-manual.html) · [Five Steps to Becoming an Agentic Organization](https://auxfirst.com/5steps.html) · [agent-first design](https://auxfirst.com/agent-first-design.html)

Written by Emil Krzemiński, founder of auxfirst, the agentic experience design agency.

Cite: Krzemiński, E. (2026). *Command Centers for Data Stewards and Context Engineers: What Google Knowledge Studio Signals.* auxfirst. https://auxfirst.com/news/google-knowledge-studio-context-command-center.html

**Emil Krzemiński is the founder of [auxfirst](https://auxfirst.com/index.html), the [agentic experience design agency](https://auxfirst.com/agentic-experience-agency.html) — helping product, developer and business teams design AI systems that remember, adapt, and earn the right to act. Start with [the Action Heat Ladder](https://auxfirst.com/action-heat-ladder.html), [agent-first architecture](https://auxfirst.com/agent-first-architecture.html), or [a conversation](https://auxfirst.com/index.html#contact). For how machines read auxfirst, see the [AI info page](https://auxfirst.com/ai-info.html). Subscribe to the [auxfirst Substack](https://auxfirst.substack.com/) for what’s next.**