# Appendix C. Preparing for CCAO-F (/appendices/preparing-for-ccao-f)

---
type: Document
title: "Appendix C. Preparing for CCAO-F"
description: "Everything the CCAO-F exam guide lists is taught in this book. This page shows where, in the exam's order, and which guide your exam follows."
status: stable
order: 903
ksor:
  owner: team:panaversity
  audience: [ public ]
  approval:
    by: process:panaversity
    at: 2026-11-03T18:48:40Z
chapter: "Appendix C"
part: Back
expert_status: required
objectives: []
word_budget: 900
prerequisites: [ "F3" ]
build_step: null
artifact: null
field_guides: []
delta_entries: []
concepts: []
last_verified: 2026-10-19
sources:
  - id: anthropic-ccao-f-guide
    title: "Claude Certified Associate: Foundations Exam Guide, version 1.0 (Anthropic, effective July 2026, verified 2026-10-19)"
    resource: "https://everpath-course-content.s3-accelerate.amazonaws.com/instructor/6nizmqk8tpzpfjvt6qmmav7rh/public/1783542847/Claude+Certified+Associate+%E2%80%93+Foundations+Exam+Guide.pdf"
  - id: anthropic-partner-certifications
    title: "Partner Certifications (Anthropic Partner Academy, verified 2026-10-17)"
    resource: https://anthropic-partners.skilljar.com/page/partner-certifications
generated:
  at: 2026-11-03T18:48:40Z
  by: process:panaversity
trust_tier: unverified
build_id: sha256:e1fbcb7d9a8f8b9b7cb96875022325c02c9ba4be94cef13b301cbd39a9edad85
dirty: true
ksor_version: 0.0.60
---

**In everyday life.** At the back of many books, an index lists each topic and the pages where you can find it. This page is an index of this book for the CCAO-F exam. It lists each exam objective and where the book teaches it.

Everything the CCAO-F exam guide lists is taught in this book. CCAO-F is Anthropic's Claude Certified Associate: Foundations exam, and its guide sets out 30 objectives in seven domains.[^anthropic-ccao-f-guide] Each objective has a lesson in this book, a lab step where you practice it, and quiz questions that check it. Every chapter in Parts I and II also has a page that shows the same work on both AI vendors.

This page shows where everything is, in the exam's order. Revise by domain, heaviest first, so your time follows the marks. If you passed PCAO-F, Panaversity's exam on Parts I and II, you have already covered every domain, and the map is your revision guide.

## Which guide your exam follows

> [!NOTE]
> This page follows guide 1.0, in force since July 2026. Anthropic publishes the next guide on November 17, 2026, and the exam follows it from December 7, 2026.[^anthropic-partner-certifications] Until then, the last part of this page shows Panaversity's projection of it. This page is revised on November 17.

![A timeline of which guide an exam follows. The date of the exam decides, not the day it is booked. Guide 1.0 is in force until Sunday, December 6, 2026. On Tuesday, November 17, Anthropic publishes the new guide. From Monday, December 7, exams follow it. An exam on Thursday, December 3 follows guide 1.0. An exam on Thursday, December 10 follows the new guide](img/appendix-c-which-guide.png)

Before you book, check the guide's version and date on Anthropic's certification page.

## The map: where the book teaches each objective

### Output Evaluation and Validation, 21 percent

| You can | Read | Practice |
| --- | --- | --- |
| Evaluate outputs for accuracy and completeness | [6.2](../managing-ai-workers/the-review-contract/good-and-plausible.md) | [Lab 6, Task 3](../managing-ai-workers/the-review-contract/lab.md#task-3-all-15-invoices-10-minutes) and [Lab 6, Task 4](../managing-ai-workers/the-review-contract/lab.md#task-4-any-bill-twice-10-minutes) |
| Spot hallucination, inconsistency and bias | [6.3](../managing-ai-workers/the-review-contract/three-failures.md) | [Lab 6, Task 2](../managing-ai-workers/the-review-contract/lab.md#task-2-what-did-it-do-15-minutes), [Lab 6, Task 7](../managing-ai-workers/the-review-contract/lab.md#task-7-do-the-files-agree-15-minutes) and [Lab 6, Task 8](../managing-ai-workers/the-review-contract/lab.md#task-8-move-northern-maple-10-minutes) |
| Fact-check and validate | [6.4](../managing-ai-workers/the-review-contract/decision-numbers.md), [6.5](../managing-ai-workers/the-review-contract/citation-and-abstention.md) and [0.2.2](../how-ai-works/your-first-hour/the-controls.md) | [Lab 6, Task 5](../managing-ai-workers/the-review-contract/lab.md#task-5-do-the-sections-hold-up-15-minutes), [Lab 6, Task 6](../managing-ai-workers/the-review-contract/lab.md#task-6-the-number-i-sign-20-minutes), [Lab 6, Task 11](../managing-ai-workers/the-review-contract/lab.md#task-11-a-second-reviewer-20-minutes) and [Lab 0.2, Task 2](../how-ai-works/your-first-hour/lab.md#task-2-answer-from-a-file) |
| Decide when human review or more verification is required | [10.7](../managing-ai-workers/governance-and-responsible-use/when-a-human-must-sign.md) and [6.1](../managing-ai-workers/the-review-contract/agree-the-checks.md) | [Lab 10, Task 8](../managing-ai-workers/governance-and-responsible-use/lab.md#task-8-who-signs-20-minutes), [Lab 10, Task 11](../managing-ai-workers/governance-and-responsible-use/lab.md#task-11-the-record-and-the-policy-20-minutes), [Lab 6, Task 1](../managing-ai-workers/the-review-contract/lab.md#task-1-the-checks-before-you-look-20-minutes) and [Lab 6, Task 9](../managing-ai-workers/the-review-contract/lab.md#task-9-approve-fix-or-return-10-minutes) |
| Edit, adapt and compare outputs for an audience | [6.6](../managing-ai-workers/the-review-contract/audience.md) | [Lab 6, Task 7](../managing-ai-workers/the-review-contract/lab.md#task-7-do-the-files-agree-15-minutes) and [Lab 6, Task 10](../managing-ai-workers/the-review-contract/lab.md#task-10-three-lines-for-linda-15-minutes) |
| Organize information and choose an output format: inline, artifact or structured data | [5.7](../managing-ai-workers/the-four-part-brief/output-format.md) | [Lab 5, Task 5](../managing-ai-workers/the-four-part-brief/lab.md#task-5-the-format-10-minutes), [Lab 5, Task 7](../managing-ai-workers/the-four-part-brief/lab.md#task-7-change-the-part-that-failed-20-minutes) and [Lab 5, Task 10](../managing-ai-workers/the-four-part-brief/lab.md#task-10-three-new-vendors-as-json-10-minutes) |

Final rounds: the Check yourself pages of [Chapter 6](../managing-ai-workers/the-review-contract/check-yourself.md), [Chapter 5](../managing-ai-workers/the-four-part-brief/check-yourself.md), [Chapter 10](../managing-ai-workers/governance-and-responsible-use/check-yourself.md) and [Chapter 0.2](../how-ai-works/your-first-hour/check-yourself.md).

Exam notes: [Lab 5](../managing-ai-workers/the-four-part-brief/lab.md#exam-notes), on features.

### Workflow Integration and Solution Design, 16 percent

| You can | Read | Practice |
| --- | --- | --- |
| Analyze requirements and use cases | [9.1](../managing-ai-workers/work-that-runs-without-you/is-this-an-agent-problem.md) | [Lab 9, Task 1](../managing-ai-workers/work-that-runs-without-you/lab.md#task-1-what-our-contract-decided-25-minutes) |
| Use AI for research, planning and process improvement | [9.2](../managing-ai-workers/work-that-runs-without-you/fit-the-workflow.md), [2.4](../ai-worker-paradigm/what-is-an-ai-worker/surface-and-model.md) and [0.2.2](../how-ai-works/your-first-hour/the-controls.md) | [Lab 9, Task 2](../managing-ai-workers/work-that-runs-without-you/lab.md#task-2-who-does-each-step-10-minutes), [Lab 9, Task 3](../managing-ai-workers/work-that-runs-without-you/lab.md#task-3-what-every-run-is-given-20-minutes) and [Lab 6, Task 12](../managing-ai-workers/the-review-contract/lab.md#task-12-research-on-your-own-work-optional-10-minutes) |
| Support solution design, development and iteration | [3.2](../ai-worker-paradigm/the-10-80-10-operating-rhythm/first-10-percent.md), [3.4](../ai-worker-paradigm/the-10-80-10-operating-rhythm/final-10-percent.md) and [9.3](../managing-ai-workers/work-that-runs-without-you/scheduled-and-event-triggered.md) | [Lab 9, Task 3](../managing-ai-workers/work-that-runs-without-you/lab.md#task-3-what-every-run-is-given-20-minutes), [Lab 9, Task 4](../managing-ai-workers/work-that-runs-without-you/lab.md#task-4-read-the-record-first-20-minutes), [Lab 9, Task 5](../managing-ai-workers/work-that-runs-without-you/lab.md#task-5-review-it-as-you-do-every-thursday-15-minutes) and [Lab 3, Task 8](../ai-worker-paradigm/the-10-80-10-operating-rhythm/lab.md#task-8-draft-2-30-minutes) |
| Integrate AI into existing workflows, or redesign them | [9.2](../managing-ai-workers/work-that-runs-without-you/fit-the-workflow.md) and [7.3](../managing-ai-workers/the-authority-envelope/choose-the-rung.md) | [Lab 9, Task 2](../managing-ai-workers/work-that-runs-without-you/lab.md#task-2-who-does-each-step-10-minutes) and [Lab 7, Task 3](../managing-ai-workers/the-authority-envelope/lab.md#task-3-how-high-it-may-go-15-minutes) |
| Explain value and limits to stakeholders | [9.7](../managing-ai-workers/work-that-runs-without-you/explaining-value-and-limits.md) | [Lab 9, Task 10](../managing-ai-workers/work-that-runs-without-you/lab.md#task-10-one-page-for-dave-15-minutes) |

Final rounds: the Check yourself pages of [Chapter 9](../managing-ai-workers/work-that-runs-without-you/check-yourself.md), [Chapter 2](../ai-worker-paradigm/what-is-an-ai-worker/check-yourself.md), [Chapter 3](../ai-worker-paradigm/the-10-80-10-operating-rhythm/check-yourself.md) and [Chapter 0.4](../how-ai-works/ai-fluency/check-yourself.md).

Exam notes: [Lab 9](../managing-ai-workers/work-that-runs-without-you/lab.md#exam-notes), on work that repeats.

### Governance, Risk, and Responsible Use, 15 percent

| You can | Read | Practice |
| --- | --- | --- |
| Tell appropriate from inappropriate use cases | [10.1](../managing-ai-workers/governance-and-responsible-use/appropriate-use.md) | [Lab 10, Task 2](../managing-ai-workers/governance-and-responsible-use/lab.md#task-2-six-proposed-uses-25-minutes) |
| Apply data sensitivity, regulation and privacy | [10.2](../managing-ai-workers/governance-and-responsible-use/can-this-data-go-in.md), [10.3](../managing-ai-workers/governance-and-responsible-use/the-route-matters.md) and [4.10](../ai-worker-paradigm/the-architecture-in-one-picture/when-others-use-it.md) | [Lab 10, Task 3](../managing-ai-workers/governance-and-responsible-use/lab.md#task-3-what-the-check-needs-15-minutes), [Lab 10, Task 4](../managing-ai-workers/governance-and-responsible-use/lab.md#task-4-which-way-in-10-minutes) and [Lab 4, Task 6](../ai-worker-paradigm/the-architecture-in-one-picture/lab.md#task-6-the-move-and-when-others-use-it-15-minutes) |
| Follow organizational AI policies | [10.4](../managing-ai-workers/governance-and-responsible-use/organizational-policy.md) and [10.9](../managing-ai-workers/governance-and-responsible-use/both-ai-vendors.md) | [Lab 10, Task 11](../managing-ai-workers/governance-and-responsible-use/lab.md#task-11-the-record-and-the-policy-20-minutes) and [Lab 10, Task 12](../managing-ai-workers/governance-and-responsible-use/lab.md#task-12-which-settings-carry-it-out-10-minutes) |
| Understand ethical implications | [10.5](../managing-ai-workers/governance-and-responsible-use/who-is-affected.md) and [0.4.2](../how-ai-works/ai-fluency/the-four-ds.md) | [Lab 10, Task 5](../managing-ai-workers/governance-and-responsible-use/lab.md#task-5-who-the-list-left-out-10-minutes), [Lab 10, Task 6](../managing-ai-workers/governance-and-responsible-use/lab.md#task-6-a-standing-brief-10-minutes) and [Lab 0.4, Step 2](../how-ai-works/ai-fluency/lab.md#step-2-decide-the-4d-plan) |

Final rounds: the Check yourself pages of [Chapter 10](../managing-ai-workers/governance-and-responsible-use/check-yourself.md), [Chapter 4](../ai-worker-paradigm/the-architecture-in-one-picture/check-yourself.md) and [Chapter 0.4](../how-ai-works/ai-fluency/check-yourself.md).

Exam notes: [Lab 10](../managing-ai-workers/governance-and-responsible-use/lab.md#exam-notes), on personal data.

### Prompting and Task Execution, 14 percent

| You can | Read | Practice |
| --- | --- | --- |
| Write effective prompts for business and technical tasks | [5.1](../managing-ai-workers/the-four-part-brief/four-parts.md), [5.2](../managing-ai-workers/the-four-part-brief/outcome-not-steps.md) and [5.3](../managing-ai-workers/the-four-part-brief/governed-source.md) | [Lab 5, Task 3](../managing-ai-workers/the-four-part-brief/lab.md#task-3-the-outcome-10-minutes), [Lab 5, Task 4](../managing-ai-workers/the-four-part-brief/lab.md#task-4-the-context-15-minutes), [Lab 5, Task 5](../managing-ai-workers/the-four-part-brief/lab.md#task-5-the-format-10-minutes), [Lab 5, Task 6](../managing-ai-workers/the-four-part-brief/lab.md#task-6-the-constraints-15-minutes), [Lab 5, Task 9](../managing-ai-workers/the-four-part-brief/lab.md#task-9-ideas-for-the-ap-inbox-10-minutes) and [Lab 5, Task 10](../managing-ai-workers/the-four-part-brief/lab.md#task-10-three-new-vendors-as-json-10-minutes) |
| Decompose complex requests | [5.4](../managing-ai-workers/the-four-part-brief/decompose.md) | [Lab 5, Task 8](../managing-ai-workers/the-four-part-brief/lab.md#task-8-where-it-stops-15-minutes), [Lab 5, Task 11](../managing-ai-workers/the-four-part-brief/lab.md#task-11-the-same-brief-a-fresh-chat-30-minutes) and [Lab 5, Task 12](../managing-ai-workers/the-four-part-brief/lab.md#task-12-helpers-20-minutes) |
| Iterate prompts to improve output | [5.5](../managing-ai-workers/the-four-part-brief/iterate.md) and [11.5](../managing-ai-workers/when-the-worker-goes-wrong/three-levers.md) | [Lab 5, Task 7](../managing-ai-workers/the-four-part-brief/lab.md#task-7-change-the-part-that-failed-20-minutes), [Lab 11, Task 5](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-5-jordans-replies-15-minutes) and [Lab 11, Task 7](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-7-two-changes-one-at-a-time-15-minutes) |
| Adapt prompting to the task type: analysis, research, drafting or brainstorming | [5.6](../managing-ai-workers/the-four-part-brief/task-type.md) | [Lab 5, Task 9](../managing-ai-workers/the-four-part-brief/lab.md#task-9-ideas-for-the-ap-inbox-10-minutes) and [Lab 5, Task 10](../managing-ai-workers/the-four-part-brief/lab.md#task-10-three-new-vendors-as-json-10-minutes) |

Final rounds: the Check yourself pages of [Chapter 5](../managing-ai-workers/the-four-part-brief/check-yourself.md) and [Chapter 11](../managing-ai-workers/when-the-worker-goes-wrong/check-yourself.md).

Exam notes: [Lab 5](../managing-ai-workers/the-four-part-brief/lab.md#exam-notes), on what a good prompt contains.

### Product and Model Selection, 12 percent

| You can | Read | Practice |
| --- | --- | --- |
| Choose the right product feature: projects, research mode, chat or artifacts | [2.4](../ai-worker-paradigm/what-is-an-ai-worker/surface-and-model.md), [2.5](../ai-worker-paradigm/what-is-an-ai-worker/both-ai-vendors.md) and [0.2.2](../how-ai-works/your-first-hour/the-controls.md) | [Lab 2, Task 5](../ai-worker-paradigm/what-is-an-ai-worker/lab.md#task-5-how-it-runs-and-is-reached-20-minutes) and [Lab 1, Task 6](../ai-worker-paradigm/from-chatbots-to-ai-workers/lab.md#task-6-which-surface-for-which-job-10-minutes) |
| Tell the model families apart: Haiku, Sonnet and Opus | [2.5](../ai-worker-paradigm/what-is-an-ai-worker/both-ai-vendors.md) and [0.1.4](../how-ai-works/what-ai-actually-is/both-ai-vendors.md) | [Lab 2, Task 5](../ai-worker-paradigm/what-is-an-ai-worker/lab.md#task-5-how-it-runs-and-is-reached-20-minutes) |
| Match the model to cost, speed and quality | [2.4](../ai-worker-paradigm/what-is-an-ai-worker/surface-and-model.md) and [2.5](../ai-worker-paradigm/what-is-an-ai-worker/both-ai-vendors.md) | [Lab 2, Task 5](../ai-worker-paradigm/what-is-an-ai-worker/lab.md#task-5-how-it-runs-and-is-reached-20-minutes) and [Lab 0.2, Task 5](../how-ai-works/your-first-hour/lab.md#task-5-a-stronger-model) |
| Manage context limits and memory: restart, summarize or persist | [8.3](../managing-ai-workers/context-memory-knowledge-and-state/restart-summarize-persist.md) and [8.2](../managing-ai-workers/context-memory-knowledge-and-state/context-window.md) | [Lab 8, Task 2](../managing-ai-workers/context-memory-knowledge-and-state/lab.md#task-2-the-long-chat-10-minutes) |

Final rounds: the Check yourself pages of [Chapter 2](../ai-worker-paradigm/what-is-an-ai-worker/check-yourself.md), [Chapter 8](../managing-ai-workers/context-memory-knowledge-and-state/check-yourself.md), [Chapter 0.1](../how-ai-works/what-ai-actually-is/check-yourself.md) and [Chapter 0.2](../how-ai-works/your-first-hour/check-yourself.md).

Exam notes: [Lab 2](../ai-worker-paradigm/what-is-an-ai-worker/lab.md#exam-notes), on model families, features and choosing a model. [Lab 8](../managing-ai-workers/context-memory-knowledge-and-state/lab.md#exam-notes), on long conversations.

### Configuration and Knowledge Management, 12 percent

| You can | Read | Practice |
| --- | --- | --- |
| Configure projects with instructions and knowledge | [8.5](../managing-ai-workers/context-memory-knowledge-and-state/configure-the-workspace.md), [8.8](../managing-ai-workers/context-memory-knowledge-and-state/both-ai-vendors.md) and [0.2.2](../how-ai-works/your-first-hour/the-controls.md) | [Lab 8, Task 5](../managing-ai-workers/context-memory-knowledge-and-state/lab.md#task-5-approved-and-what-the-project-loads-15-minutes), [Lab 8, Task 13](../managing-ai-workers/context-memory-knowledge-and-state/lab.md#task-13-nothing-but-the-records-20-minutes) and [Lab 0.2, Task 6](../how-ai-works/your-first-hour/lab.md#task-6-make-a-project) |
| Manage uploaded knowledge and connectors, such as Google Drive and Gmail | [8.5](../managing-ai-workers/context-memory-knowledge-and-state/configure-the-workspace.md) and [8.8](../managing-ai-workers/context-memory-knowledge-and-state/both-ai-vendors.md) | [Lab 8, Task 5](../managing-ai-workers/context-memory-knowledge-and-state/lab.md#task-5-approved-and-what-the-project-loads-15-minutes), [Lab 8, Task 10](../managing-ai-workers/context-memory-knowledge-and-state/lab.md#task-10-one-change-every-copy-25-minutes) and [Lab 8, Task 12](../managing-ai-workers/context-memory-knowledge-and-state/lab.md#task-12-what-it-holds-and-reaches-15-minutes) |
| Write system-level instructions | [8.5](../managing-ai-workers/context-memory-knowledge-and-state/configure-the-workspace.md) and [8.8](../managing-ai-workers/context-memory-knowledge-and-state/both-ai-vendors.md) | [Lab 8, Task 6](../managing-ai-workers/context-memory-knowledge-and-state/lab.md#task-6-standing-instructions-15-minutes) |
| Maintain and update configurations, knowledge and instructions | [8.7](../managing-ai-workers/context-memory-knowledge-and-state/knowledge-owner.md), [8.5](../managing-ai-workers/context-memory-knowledge-and-state/configure-the-workspace.md) and [8.8](../managing-ai-workers/context-memory-knowledge-and-state/both-ai-vendors.md) | [Lab 8, Task 10](../managing-ai-workers/context-memory-knowledge-and-state/lab.md#task-10-one-change-every-copy-25-minutes) |

Final rounds: the Check yourself pages of [Chapter 8](../managing-ai-workers/context-memory-knowledge-and-state/check-yourself.md) and [Chapter 0.2](../how-ai-works/your-first-hour/check-yourself.md).

Exam notes: [Lab 8](../managing-ai-workers/context-memory-knowledge-and-state/lab.md#exam-notes), on project knowledge and connectors.

### Troubleshooting and Optimization, 10 percent

| You can | Read | Practice |
| --- | --- | --- |
| Diagnose and fix underperforming prompts and outputs | [11.3](../managing-ai-workers/when-the-worker-goes-wrong/where-a-failure-lives.md) and [11.2](../managing-ai-workers/when-the-worker-goes-wrong/ran-versus-worked.md) | [Lab 11, Task 5](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-5-jordans-replies-15-minutes), [Lab 11, Task 8](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-8-the-first-error-and-the-access-20-minutes), [Lab 11, Task 10](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-10-the-discounts-15-minutes), [Lab 9, Task 1](../managing-ai-workers/work-that-runs-without-you/lab.md#task-1-what-our-contract-decided-25-minutes) and [Lab 9, Task 5](../managing-ai-workers/work-that-runs-without-you/lab.md#task-5-review-it-as-you-do-every-thursday-15-minutes) |
| Adjust based on feedback and results | [11.6](../managing-ai-workers/when-the-worker-goes-wrong/make-the-fix-stick.md) and [11.5](../managing-ai-workers/when-the-worker-goes-wrong/three-levers.md) | [Lab 11, Task 11](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-11-where-each-correction-lives-20-minutes), [Lab 11, Task 4](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-4-tuesdays-two-fixes-25-minutes) and [Lab 11, Task 13](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-13-the-same-week-again-15-minutes) |
| Optimize workflows | [11.6](../managing-ai-workers/when-the-worker-goes-wrong/make-the-fix-stick.md) and [9.2](../managing-ai-workers/work-that-runs-without-you/fit-the-workflow.md) | [Lab 11, Task 11](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-11-where-each-correction-lives-20-minutes) and [Lab 9, Task 2](../managing-ai-workers/work-that-runs-without-you/lab.md#task-2-who-does-each-step-10-minutes) |

Final rounds: the Check yourself pages of [Chapter 11](../managing-ai-workers/when-the-worker-goes-wrong/check-yourself.md) and [Chapter 9](../managing-ai-workers/work-that-runs-without-you/check-yourself.md).

Exam notes: [Lab 11](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#exam-notes), on fixing a weak output.

## Check that you are ready

Each page ends with a short quiz, and each chapter ends with a final round on its Check yourself page. When you finish a round, the results show your score for each exam domain, as the CCAO-F score report does.[^anthropic-ccao-f-guide] If a domain scores low, go back to its rows in the map. Then answer the three sample questions in the guide. When you are ready for the clock, sit the [PCAO-F sample exam](pcao-f-sample-exam.md): 60 questions in 120 minutes, in the CCAO-F format.

## If you sit the exam on or after December 7, 2026

> [!NOTE]
> This part is Panaversity's projection, not Anthropic's guide. The published guide replaces it on November 17, 2026.

The projection adds these objectives to the ones in the map.

| Domain | You can | Read | Practice |
| --- | --- | --- | --- |
| Prompting and Task Execution | Brief an agent for a multi-step task: goal, constraints, checkpoints | [5.1](../managing-ai-workers/the-four-part-brief/four-parts.md) and [5.4](../managing-ai-workers/the-four-part-brief/decompose.md) | [Lab 5, Task 3](../managing-ai-workers/the-four-part-brief/lab.md#task-3-the-outcome-10-minutes), [Lab 5, Task 4](../managing-ai-workers/the-four-part-brief/lab.md#task-4-the-context-15-minutes), [Lab 5, Task 5](../managing-ai-workers/the-four-part-brief/lab.md#task-5-the-format-10-minutes), [Lab 5, Task 6](../managing-ai-workers/the-four-part-brief/lab.md#task-6-the-constraints-15-minutes) and [Lab 5, Task 8](../managing-ai-workers/the-four-part-brief/lab.md#task-8-where-it-stops-15-minutes) |
|  | Write reusable instructions as a skill instead of prompting again | [8.9](../managing-ai-workers/context-memory-knowledge-and-state/skills-and-plugins.md) and [8.5](../managing-ai-workers/context-memory-knowledge-and-state/configure-the-workspace.md) | [Lab 8, Task 9](../managing-ai-workers/context-memory-knowledge-and-state/lab.md#task-9-fix-it-and-who-approves-20-minutes) |
| Output Evaluation and Validation | Review the output of long-running or agentic work before acting on it | [6.7](../managing-ai-workers/the-review-contract/work-you-did-not-watch.md), [6.8](../managing-ai-workers/the-review-contract/both-ai-vendors.md) and [11.1](../managing-ai-workers/when-the-worker-goes-wrong/stop-first.md) | [Lab 6, Task 2](../managing-ai-workers/the-review-contract/lab.md#task-2-what-did-it-do-15-minutes), [Lab 6, Task 11](../managing-ai-workers/the-review-contract/lab.md#task-11-a-second-reviewer-20-minutes), [Lab 11, Task 6](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-6-ran-or-worked-10-minutes) and [Lab 11, Task 8](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-8-the-first-error-and-the-access-20-minutes) |
|  | Validate output that cites live web sources or connector data | [6.5](../managing-ai-workers/the-review-contract/citation-and-abstention.md), [6.8](../managing-ai-workers/the-review-contract/both-ai-vendors.md) and [8.5](../managing-ai-workers/context-memory-knowledge-and-state/configure-the-workspace.md) | [Lab 6, Task 12](../managing-ai-workers/the-review-contract/lab.md#task-12-research-on-your-own-work-optional-10-minutes) |
|  | Check a research report's sources before relying on it | [6.5](../managing-ai-workers/the-review-contract/citation-and-abstention.md) | [Lab 6, Task 12](../managing-ai-workers/the-review-contract/lab.md#task-12-research-on-your-own-work-optional-10-minutes) |
|  | Check computed results from uploaded data files against the source data | [6.4](../managing-ai-workers/the-review-contract/decision-numbers.md) and [0.2.2](../how-ai-works/your-first-hour/the-controls.md) | [Lab 6, Task 6](../managing-ai-workers/the-review-contract/lab.md#task-6-the-number-i-sign-20-minutes), [Lab 0.2, Task 4](../how-ai-works/your-first-hour/lab.md#task-4-let-it-run-code) and [Lab 11, Task 10](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-10-the-discounts-15-minutes) |
|  | Choose among typed deliverables (slides, documents, sheets, dashboards) for the audience | [5.7](../managing-ai-workers/the-four-part-brief/output-format.md) | None yet |
|  | Tell a proposed action, an attempted action and a confirmed external change apart | [6.7](../managing-ai-workers/the-review-contract/work-you-did-not-watch.md), [11.1](../managing-ai-workers/when-the-worker-goes-wrong/stop-first.md) and [4.10](../ai-worker-paradigm/the-architecture-in-one-picture/when-others-use-it.md) | [Lab 7, Task 10](../managing-ai-workers/the-authority-envelope/lab.md#task-10-what-really-changed-10-minutes), [Lab 0.2, Task 8](../how-ai-works/your-first-hour/lab.md#task-8-sort-what-you-made) and [Lab 11, Task 2](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-2-stop-and-hold-15-minutes) |
| Product, Model, and Surface Selection | Tell the four tiers apart and match the tier to the task's economics | [2.4](../ai-worker-paradigm/what-is-an-ai-worker/surface-and-model.md), [2.5](../ai-worker-paradigm/what-is-an-ai-worker/both-ai-vendors.md) and [0.1.4](../how-ai-works/what-ai-actually-is/both-ai-vendors.md) | [Lab 2, Task 5](../ai-worker-paradigm/what-is-an-ai-worker/lab.md#task-5-how-it-runs-and-is-reached-20-minutes) |
|  | Choose where to use Claude for a workflow: web, desktop, Chrome, Microsoft 365, Slack | [1.5](../ai-worker-paradigm/from-chatbots-to-ai-workers/both-ai-vendors.md), [1.4](../ai-worker-paradigm/from-chatbots-to-ai-workers/continuity-execution-persistence.md), [2.5](../ai-worker-paradigm/what-is-an-ai-worker/both-ai-vendors.md) and [2.3](../ai-worker-paradigm/what-is-an-ai-worker/worker-runtime-channel.md) | [Lab 1, Task 8](../ai-worker-paradigm/from-chatbots-to-ai-workers/lab.md#task-8-save-it-where-you-will-find-it-monday-15-to-25-minutes), [Lab 1, Task 14](../ai-worker-paradigm/from-chatbots-to-ai-workers/lab.md#task-14-the-ai-inside-the-spreadsheet-15-minutes) and [Lab 2, Task 5](../ai-worker-paradigm/what-is-an-ai-worker/lab.md#task-5-how-it-runs-and-is-reached-20-minutes) |
|  | Configure and audit memory: chat memory, project memory, incognito chats | [8.4](../managing-ai-workers/context-memory-knowledge-and-state/memory-and-ssor.md), [8.8](../managing-ai-workers/context-memory-knowledge-and-state/both-ai-vendors.md), [0.2.1](../how-ai-works/your-first-hour/where-it-goes.md) and [0.2.3](../how-ai-works/your-first-hour/both-ai-vendors.md) | [Lab 8, Task 3](../managing-ai-workers/context-memory-knowledge-and-state/lab.md#task-3-what-memory-may-keep-10-minutes), [Lab 0.2, Task 7](../how-ai-works/your-first-hour/lab.md#task-7-memory-and-a-temporary-chat) and [Lab 0.2, Task 1](../how-ai-works/your-first-hour/lab.md#task-1-make-the-two-choices) |
|  | Choose the research depth: a quick search, deep research or your own connected data | [0.2.2](../how-ai-works/your-first-hour/the-controls.md) and [2.4](../ai-worker-paradigm/what-is-an-ai-worker/surface-and-model.md) | None yet |
| Workflow Integration and Agentic Solution Design | Design a workflow that hands steps to Claude as an agent | [9.2](../managing-ai-workers/work-that-runs-without-you/fit-the-workflow.md), [3.1](../ai-worker-paradigm/the-10-80-10-operating-rhythm/one-rhythm.md) and [9.1](../managing-ai-workers/work-that-runs-without-you/is-this-an-agent-problem.md) | [Lab 9, Task 2](../managing-ai-workers/work-that-runs-without-you/lab.md#task-2-who-does-each-step-10-minutes), [Lab 9, Task 3](../managing-ai-workers/work-that-runs-without-you/lab.md#task-3-what-every-run-is-given-20-minutes) and [Lab 3, Task 4](../ai-worker-paradigm/the-10-80-10-operating-rhythm/lab.md#task-4-the-review-contract-15-minutes) |
|  | Set up and supervise scheduled and recurring tasks | [9.3](../managing-ai-workers/work-that-runs-without-you/scheduled-and-event-triggered.md), [3.7](../ai-worker-paradigm/the-10-80-10-operating-rhythm/both-ai-vendors.md) and [9.8](../managing-ai-workers/work-that-runs-without-you/both-ai-vendors.md) | [Lab 9, Task 8](../managing-ai-workers/work-that-runs-without-you/lab.md#task-8-whose-name-and-who-notices-15-minutes), [Lab 9, Task 9](../managing-ai-workers/work-that-runs-without-you/lab.md#task-9-on-each-ai-vendor-10-minutes) and [Lab 3, Task 9](../ai-worker-paradigm/the-10-80-10-operating-rhythm/lab.md#task-9-limits-on-a-schedule-15-minutes) |
|  | Decide whether to delegate or do the work, and set human checkpoints | [3.1](../ai-worker-paradigm/the-10-80-10-operating-rhythm/one-rhythm.md) and [0.4.2](../how-ai-works/ai-fluency/the-four-ds.md) | [Lab 0.4, Step 2](../how-ai-works/ai-fluency/lab.md#step-2-decide-the-4d-plan), [Lab 3, Task 4](../ai-worker-paradigm/the-10-80-10-operating-rhythm/lab.md#task-4-the-review-contract-15-minutes) and [Lab 9, Task 2](../managing-ai-workers/work-that-runs-without-you/lab.md#task-2-who-does-each-step-10-minutes) |
|  | Work with Claude where teammates see and build on the same work | [9.9](../managing-ai-workers/work-that-runs-without-you/shared-room.md), [2.3](../ai-worker-paradigm/what-is-an-ai-worker/worker-runtime-channel.md) and [4.10](../ai-worker-paradigm/the-architecture-in-one-picture/when-others-use-it.md) | [Lab 9, Task 11](../managing-ai-workers/work-that-runs-without-you/lab.md#task-11-who-may-ask-20-minutes) and [Lab 4, Task 6](../ai-worker-paradigm/the-architecture-in-one-picture/lab.md#task-6-the-move-and-when-others-use-it-15-minutes) |
| Configuration and Knowledge Management | Find, evaluate and install plugins and connectors | [8.9](../managing-ai-workers/context-memory-knowledge-and-state/skills-and-plugins.md), [7.4](../managing-ai-workers/the-authority-envelope/blast-radius.md) and [8.8](../managing-ai-workers/context-memory-knowledge-and-state/both-ai-vendors.md) | [Lab 8, Task 8](../managing-ai-workers/context-memory-knowledge-and-state/lab.md#task-8-a-skill-from-outside-15-minutes) |
|  | Create and maintain skills for recurring work | [8.9](../managing-ai-workers/context-memory-knowledge-and-state/skills-and-plugins.md) and [8.5](../managing-ai-workers/context-memory-knowledge-and-state/configure-the-workspace.md) | [Lab 8, Task 8](../managing-ai-workers/context-memory-knowledge-and-state/lab.md#task-8-a-skill-from-outside-15-minutes), [Lab 8, Task 9](../managing-ai-workers/context-memory-knowledge-and-state/lab.md#task-9-fix-it-and-who-approves-20-minutes) and [Lab 8, Task 10](../managing-ai-workers/context-memory-knowledge-and-state/lab.md#task-10-one-change-every-copy-25-minutes) |
|  | Manage project knowledge, instructions and memory together | [8.4](../managing-ai-workers/context-memory-knowledge-and-state/memory-and-ssor.md), [8.5](../managing-ai-workers/context-memory-knowledge-and-state/configure-the-workspace.md), [8.8](../managing-ai-workers/context-memory-knowledge-and-state/both-ai-vendors.md) and [8.3](../managing-ai-workers/context-memory-knowledge-and-state/restart-summarize-persist.md) | [Lab 8, Task 3](../managing-ai-workers/context-memory-knowledge-and-state/lab.md#task-3-what-memory-may-keep-10-minutes), [Lab 8, Task 5](../managing-ai-workers/context-memory-knowledge-and-state/lab.md#task-5-approved-and-what-the-project-loads-15-minutes) and [Lab 8, Task 6](../managing-ai-workers/context-memory-knowledge-and-state/lab.md#task-6-standing-instructions-15-minutes) |
| Governance, Risk, and Responsible Use | Apply the 2026 Usage Policy | [10.9](../managing-ai-workers/governance-and-responsible-use/both-ai-vendors.md) and [10.3](../managing-ai-workers/governance-and-responsible-use/the-route-matters.md) | None yet |
|  | Set autonomy and approval limits for agentic actions | [7.2](../managing-ai-workers/the-authority-envelope/autonomy-ladder.md), [7.3](../managing-ai-workers/the-authority-envelope/choose-the-rung.md) and [7.1](../managing-ai-workers/the-authority-envelope/envelope-brief-permissions.md) | [Lab 7, Task 2](../managing-ai-workers/the-authority-envelope/lab.md#task-2-what-it-may-do-15-minutes), [Lab 7, Task 3](../managing-ai-workers/the-authority-envelope/lab.md#task-3-how-high-it-may-go-15-minutes) and [Lab 7, Task 4](../managing-ai-workers/the-authority-envelope/lab.md#task-4-the-numbers-15-minutes) |
|  | Judge data exposure through connectors and plugins | [7.4](../managing-ai-workers/the-authority-envelope/blast-radius.md) and [10.10](../managing-ai-workers/governance-and-responsible-use/who-can-open-the-result.md) | [Lab 7, Task 6](../managing-ai-workers/the-authority-envelope/lab.md#task-6-two-add-ons-10-minutes) |
|  | Recognize prompt injection: retrieved content never authorizes an action | [7.6](../managing-ai-workers/the-authority-envelope/untrusted-input.md) and [5.3](../managing-ai-workers/the-four-part-brief/governed-source.md) | [Lab 7, Task 9](../managing-ai-workers/the-authority-envelope/lab.md#task-9-break-the-overlap-10-minutes), [Lab 7, Task 12](../managing-ai-workers/the-authority-envelope/lab.md#task-12-this-mornings-inbox-25-minutes), [Lab 5, Task 4](../managing-ai-workers/the-four-part-brief/lab.md#task-4-the-context-15-minutes) and [Lab 5, Task 7](../managing-ai-workers/the-four-part-brief/lab.md#task-7-change-the-part-that-failed-20-minutes) |
|  | Give the least access a run needs (read, change, send) before it starts | [7.4](../managing-ai-workers/the-authority-envelope/blast-radius.md) and [7.8](../managing-ai-workers/the-authority-envelope/both-ai-vendors.md) | [Lab 7, Task 1](../managing-ai-workers/the-authority-envelope/lab.md#task-1-what-let-it-20-minutes), [Lab 7, Task 6](../managing-ai-workers/the-authority-envelope/lab.md#task-6-two-add-ons-10-minutes) and [Lab 7, Task 8](../managing-ai-workers/the-authority-envelope/lab.md#task-8-can-no-wider-than-may-15-minutes) |
|  | Judge what is safe to publish or share outward | [10.10](../managing-ai-workers/governance-and-responsible-use/who-can-open-the-result.md) and [10.9](../managing-ai-workers/governance-and-responsible-use/both-ai-vendors.md) | [Lab 10, Task 10](../managing-ai-workers/governance-and-responsible-use/lab.md#task-10-who-can-open-the-result-15-minutes) and [Lab 10, Task 11](../managing-ai-workers/governance-and-responsible-use/lab.md#task-11-the-record-and-the-policy-20-minutes) |
| Troubleshooting and Optimization | Find why an agentic task stalled or gave poor output | [11.3](../managing-ai-workers/when-the-worker-goes-wrong/where-a-failure-lives.md), [11.2](../managing-ai-workers/when-the-worker-goes-wrong/ran-versus-worked.md) and [11.1](../managing-ai-workers/when-the-worker-goes-wrong/stop-first.md) | [Lab 11, Task 3](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-3-why-it-never-started-15-minutes), [Lab 11, Task 6](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-6-ran-or-worked-10-minutes), [Lab 11, Task 12](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-12-three-more-records-20-minutes), [Lab 9, Task 1](../managing-ai-workers/work-that-runs-without-you/lab.md#task-1-what-our-contract-decided-25-minutes) and [Lab 9, Task 5](../managing-ai-workers/work-that-runs-without-you/lab.md#task-5-review-it-as-you-do-every-thursday-15-minutes) |
|  | Choose to fix the prompt, the skill or the knowledge source, or to escalate | [11.5](../managing-ai-workers/when-the-worker-goes-wrong/three-levers.md) and [11.6](../managing-ai-workers/when-the-worker-goes-wrong/make-the-fix-stick.md) | [Lab 11, Task 11](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-11-where-each-correction-lives-20-minutes) and [Lab 11, Task 4](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-4-tuesdays-two-fixes-25-minutes) |
|  | Trace a scheduled task that fails to its schedule, its access or a resource that is not available | [11.3](../managing-ai-workers/when-the-worker-goes-wrong/where-a-failure-lives.md), [11.2](../managing-ai-workers/when-the-worker-goes-wrong/ran-versus-worked.md), [11.4](../managing-ai-workers/when-the-worker-goes-wrong/usage-and-limits.md) and [11.7](../managing-ai-workers/when-the-worker-goes-wrong/both-ai-vendors.md) | [Lab 11, Task 3](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-3-why-it-never-started-15-minutes), [Lab 11, Task 4](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-4-tuesdays-two-fixes-25-minutes) and [Lab 11, Task 12](../managing-ai-workers/when-the-worker-goes-wrong/lab.md#task-12-three-more-records-20-minutes) |

## Versions of this page

- **October 2026:** written for guide 1.0, with Panaversity's projection of the next guide.

[^anthropic-ccao-f-guide]: Claude Certified Associate: Foundations Exam Guide, version 1.0, Anthropic, effective July 2026.

[^anthropic-partner-certifications]: Partner Certifications, Anthropic Partner Academy.