AI & DIGITAL SKILLS MAGAZINE Learn from beginner to expert
Contact EDITION 2026
ISHRAQA7 — The practical digital learning magazine
ISHRAQA 7 ACADEMY · Artificial intelligence

AI Agents and Automation: building reliable supervised workflows

Twelve hands-on chapters on supervised workflows, deterministic triage and mock API testing without credentials.

  • LevelIntermediate → Advanced
  • Course programme12 modules
  • Course filesPDF · CSV · XLSX · JSON · TXT · PY · ZIP

Also available in Français العربية

Twelve hands-on chapters on supervised workflows, deterministic triage and mock API testing without credentials. Synthetic training cases and data; not real ISHRAQA7 performance.

Twelve-chapter curriculum

  1. Automation, assistants and agents
  2. Architecture: triggers, tools and state
  3. Scoping goals and permissions
  4. A first supervised workflow
  5. Structuring and validating data
  6. No-code platforms: n8n, Make and Zapier
  7. API connections without real secrets
  8. Introducing a model into the workflow
  9. Human oversight and error handling
  10. Testing, logs and metrics
  11. Governance, costs and privacy
  12. Capstone: supervised automation

1. Automation, assistants and agents

Workplace application · Intermediate to advanced

A fictional service receives forty non-personal requests and wants a triage table.

Method and workflow — TRACE method

A script follows explicit rules, an assistant suggests content, and an agent may orchestrate steps and tools. Greater autonomy increases compound error risk: start with a narrow scope and a human approval gate. The outcome of this chapter is observable: Scope diagram with a mandatory approval step. A professional result keeps evidence, assumptions and decision owners visible. Any assertion without accessible proof remains pending rather than being inferred from polished language.

Draw input → validation → classification → draft → human approval → archive. Label automatic, manual and stop points. Restrict every experiment to synthetic data. Check scope, input quality, transformations, output format and approval in that order. Preserve the first attempt and justify each edit. Never hide unresolved cases simply to make the presentation look complete.

Example prompt

Training scenario: A fictional service receives forty non-personal requests and wants a triage table. Goal: Scope diagram with a mandatory approval step. Use only supplied inputs; distinguish facts, assumptions and open questions. Do not cite unseen sources. Return a table with decision, reason, evidence and status, followed by unresolved checks.

Independent exercise

Work through this scenario: A fictional service receives forty non-personal requests and wants a triage table. Deliver scope diagram with a mandatory approval step. Create two versions using the same source data, check consistency, log discrepancies and name the required approving role.

Quality safeguard

Never present an agent as infallible or free to act without boundaries.

Knowledge check

1. Which principle applies to the chapter “Automation, assistants and agents”?

  1. Inputs can be omitted.
  2. Deliver: Scope diagram with a mandatory approval step.
  3. An unverified answer is official policy.
2. What should a responsible team do in a similar case?

  1. Change evidence during a comparison.
  2. Ignore missing cases for speed.
  3. Follow the method: Draw input → validation → classification → draft → human approval → archive. Label automatic, manual
3. Which control reduces professional errors?

  1. Apply this safeguard: Never present an agent as infallible or free to act without boundaries.
  2. Run without authorisation.
  3. Publish the first output immediately.

2. Architecture: triggers, tools and state

Workplace application · Intermediate to advanced

Two fictional events use the same request identifier.

Method and workflow — TRACE method

A robust automation separates trigger, transformation, decision, state and output. Unique request IDs help prevent double handling; a log makes routing decisions auditable. The outcome of this chapter is observable: Input schema and permitted state transitions. A professional result keeps evidence, assumptions and decision owners visible. Any assertion without accessible proof remains pending rather than being inferred from polished language.

Define a JSON schema with request_id, category, urgency, reason, status and approved_by. Add uniqueness validation and a review-needed state. Specify how malformed inputs are handled. Check scope, input quality, transformations, output format and approval in that order. Preserve the first attempt and justify each edit. Never hide unresolved cases simply to make the presentation look complete.

Example prompt

Training scenario: Two fictional events use the same request identifier. Goal: Input schema and permitted state transitions. Use only supplied inputs; distinguish facts, assumptions and open questions. Do not cite unseen sources. Return a table with decision, reason, evidence and status, followed by unresolved checks.

Independent exercise

Work through this scenario: Two fictional events use the same request identifier. Deliver input schema and permitted state transitions. Create two versions using the same source data, check consistency, log discrepancies and name the required approving role.

Quality safeguard

Do not mix secrets, data and configuration in logs.

Knowledge check

1. Which principle applies to the chapter “Architecture: triggers, tools and state”?

  1. An unverified answer is official policy.
  2. Inputs can be omitted.
  3. Deliver: Input schema and permitted state transitions.
2. What should a responsible team do in a similar case?

  1. Follow the method: Define a JSON schema with request_id, category, urgency, reason, status and approved_by. Add uniquen
  2. Change evidence during a comparison.
  3. Ignore missing cases for speed.
3. Which control reduces professional errors?

  1. Publish the first output immediately.
  2. Apply this safeguard: Do not mix secrets, data and configuration in logs.
  3. Run without authorisation.

3. Scoping goals and permissions

Workplace application · Intermediate to advanced

A sorting workflow unnecessarily requests access to the real customer list.

Method and workflow — TRACE method

Least privilege gives workflows only the rights they need. A document triage has no reason to hold bulk-email or deletion privileges. Explicit permission must precede every external side effect. The outcome of this chapter is observable: Permission matrix with justified denials. A professional result keeps evidence, assumptions and decision owners visible. Any assertion without accessible proof remains pending rather than being inferred from polished language.

Write an operation/permission/risk/approver matrix. Remove unnecessary operations. Use only local training files, not live application connections, in the exercise. Check scope, input quality, transformations, output format and approval in that order. Preserve the first attempt and justify each edit. Never hide unresolved cases simply to make the presentation look complete.

Example prompt

Training scenario: A sorting workflow unnecessarily requests access to the real customer list. Goal: Permission matrix with justified denials. Use only supplied inputs; distinguish facts, assumptions and open questions. Do not cite unseen sources. Return a table with decision, reason, evidence and status, followed by unresolved checks.

Independent exercise

Work through this scenario: A sorting workflow unnecessarily requests access to the real customer list. Deliver permission matrix with justified denials. Create two versions using the same source data, check consistency, log discrepancies and name the required approving role.

Quality safeguard

Never put live API tokens in training archives.

Knowledge check

1. Which principle applies to the chapter “Scoping goals and permissions”?

  1. Deliver: Permission matrix with justified denials.
  2. An unverified answer is official policy.
  3. Inputs can be omitted.
2. What should a responsible team do in a similar case?

  1. Ignore missing cases for speed.
  2. Follow the method: Write an operation/permission/risk/approver matrix. Remove unnecessary operations. Use only local tr
  3. Change evidence during a comparison.
3. Which control reduces professional errors?

  1. Run without authorisation.
  2. Publish the first output immediately.
  3. Apply this safeguard: Never put live API tokens in training archives.

4. A first supervised workflow

Workplace application · Intermediate to advanced

Forty fictional tickets need triage into information, support and complaint.

Method and workflow — TRACE method

The first effective workflow stays deterministic with transparent routing rules. An optional model may suggest a category; a person approves any outbound action or closure. The outcome of this chapter is observable: Output CSV with rationale and approval status. A professional result keeps evidence, assumptions and decision owners visible. Any assertion without accessible proof remains pending rather than being inferred from polished language.

Read the request CSV, normalise text, classify using transparent rules, and build a human-review queue. Trigger no communication. Compare input and output row counts. Check scope, input quality, transformations, output format and approval in that order. Preserve the first attempt and justify each edit. Never hide unresolved cases simply to make the presentation look complete.

Example prompt

Training scenario: Forty fictional tickets need triage into information, support and complaint. Goal: Output CSV with rationale and approval status. Use only supplied inputs; distinguish facts, assumptions and open questions. Do not cite unseen sources. Return a table with decision, reason, evidence and status, followed by unresolved checks.

Independent exercise

Work through this scenario: Forty fictional tickets need triage into information, support and complaint. Deliver output csv with rationale and approval status. Create two versions using the same source data, check consistency, log discrepancies and name the required approving role.

Quality safeguard

Route ambiguity to human review; never guess.

Knowledge check

1. Which principle applies to the chapter “A first supervised workflow”?

  1. Inputs can be omitted.
  2. Deliver: Output CSV with rationale and approval status.
  3. An unverified answer is official policy.
2. What should a responsible team do in a similar case?

  1. Change evidence during a comparison.
  2. Ignore missing cases for speed.
  3. Follow the method: Read the request CSV, normalise text, classify using transparent rules, and build a human-review que
3. Which control reduces professional errors?

  1. Apply this safeguard: Route ambiguity to human review; never guess.
  2. Run without authorisation.
  3. Publish the first output immediately.

5. Structuring and validating data

Workplace application · Intermediate to advanced

An exported file has duplicate IDs and missing text.

Method and workflow — TRACE method

AI stages cannot fix a broken input contract. Essential checks include required fields, type, allowed values, uniqueness and text length limits. The outcome of this chapter is observable: Quality report: accepted, rejected and review-needed. A professional result keeps evidence, assumptions and decision owners visible. Any assertion without accessible proof remains pending rather than being inferred from polished language.

Build a validator with required fields and explicit error codes. Test valid input, blank text, duplicate ID and unknown category. Keep source data separate from rejected records. Check scope, input quality, transformations, output format and approval in that order. Preserve the first attempt and justify each edit. Never hide unresolved cases simply to make the presentation look complete.

Example prompt

Training scenario: An exported file has duplicate IDs and missing text. Goal: Quality report: accepted, rejected and review-needed. Use only supplied inputs; distinguish facts, assumptions and open questions. Do not cite unseen sources. Return a table with decision, reason, evidence and status, followed by unresolved checks.

Independent exercise

Work through this scenario: An exported file has duplicate IDs and missing text. Deliver quality report: accepted, rejected and review-needed. Create two versions using the same source data, check consistency, log discrepancies and name the required approving role.

Quality safeguard

Never silently drop invalid records.

Knowledge check

1. Which principle applies to the chapter “Structuring and validating data”?

  1. An unverified answer is official policy.
  2. Inputs can be omitted.
  3. Deliver: Quality report: accepted, rejected and review-needed.
2. What should a responsible team do in a similar case?

  1. Follow the method: Build a validator with required fields and explicit error codes. Test valid input, blank text, dupli
  2. Change evidence during a comparison.
  3. Ignore missing cases for speed.
3. Which control reduces professional errors?

  1. Publish the first output immediately.
  2. Apply this safeguard: Never silently drop invalid records.
  3. Run without authorisation.

6. No-code platforms: n8n, Make and Zapier

Workplace application · Intermediate to advanced

A team wants to move the same process between platforms.

Method and workflow — TRACE method

No-code scenarios combine triggers and processing nodes. Exports, credentials and usage limits vary by platform; a conceptual diagram is not necessarily importable. Record precisely what was tested. The outcome of this chapter is observable: Portability table listing untested requirements. A professional result keeps evidence, assumptions and decision owners visible. Any assertion without accessible proof remains pending rather than being inferred from polished language.

Design a manual-trigger workflow for each platform: load CSV, validate, route, stop. For n8n, inspect a credential-free JSON sample and compare its structure with official documentation. Check scope, input quality, transformations, output format and approval in that order. Preserve the first attempt and justify each edit. Never hide unresolved cases simply to make the presentation look complete.

Example prompt

Training scenario: A team wants to move the same process between platforms. Goal: Portability table listing untested requirements. Use only supplied inputs; distinguish facts, assumptions and open questions. Do not cite unseen sources. Return a table with decision, reason, evidence and status, followed by unresolved checks.

Independent exercise

Work through this scenario: A team wants to move the same process between platforms. Deliver portability table listing untested requirements. Create two versions using the same source data, check consistency, log discrepancies and name the required approving role.

Quality safeguard

Never claim import compatibility without a native import test.

Knowledge check

1. Which principle applies to the chapter “No-code platforms: n8n, Make and Zapier”?

  1. Deliver: Portability table listing untested requirements.
  2. An unverified answer is official policy.
  3. Inputs can be omitted.
2. What should a responsible team do in a similar case?

  1. Ignore missing cases for speed.
  2. Follow the method: Design a manual-trigger workflow for each platform: load CSV, validate, route, stop. For n8n, inspec
  3. Change evidence during a comparison.
3. Which control reduces professional errors?

  1. Run without authorisation.
  2. Publish the first output immediately.
  3. Apply this safeguard: Never claim import compatibility without a native import test.

7. API connections without real secrets

Workplace application · Intermediate to advanced

The mock service returns 503 on two requests and 400 on malformed input.

Method and workflow — TRACE method

An API exposes a contract: endpoint, method, payload, status and error. Training can use a local simulated server, without Internet access or live secrets. The outcome of this chapter is observable: Mock request log with retries and escalation. A professional result keeps evidence, assumptions and decision owners visible. Any assertion without accessible proof remains pending rather than being inferred from polished language.

Define a mock response with status 200, 400 and 503. Add bounded timeouts and idempotent retry rules. Log only request identifiers and status codes, never confidential content. Check scope, input quality, transformations, output format and approval in that order. Preserve the first attempt and justify each edit. Never hide unresolved cases simply to make the presentation look complete.

Example prompt

Training scenario: The mock service returns 503 on two requests and 400 on malformed input. Goal: Mock request log with retries and escalation. Use only supplied inputs; distinguish facts, assumptions and open questions. Do not cite unseen sources. Return a table with decision, reason, evidence and status, followed by unresolved checks.

Independent exercise

Work through this scenario: The mock service returns 503 on two requests and 400 on malformed input. Deliver mock request log with retries and escalation. Create two versions using the same source data, check consistency, log discrepancies and name the required approving role.

Quality safeguard

Do not expose a local training API to the public Internet.

Knowledge check

1. Which principle applies to the chapter “API connections without real secrets”?

  1. Inputs can be omitted.
  2. Deliver: Mock request log with retries and escalation.
  3. An unverified answer is official policy.
2. What should a responsible team do in a similar case?

  1. Change evidence during a comparison.
  2. Ignore missing cases for speed.
  3. Follow the method: Define a mock response with status 200, 400 and 503. Add bounded timeouts and idempotent retry rules
3. Which control reduces professional errors?

  1. Apply this safeguard: Do not expose a local training API to the public Internet.
  2. Run without authorisation.
  3. Publish the first output immediately.

8. Introducing a model into the workflow

Workplace application · Intermediate to advanced

A mock classifier returns a fourth category not in the schema.

Method and workflow — TRACE method

A model stage is non-deterministic and may alter output format. Restrict it to a validated schema, route uncertainty to review, and offer a manual fallback when the model is unavailable. The outcome of this chapter is observable: JSON output contract, checks and manual fallback. A professional result keeps evidence, assumptions and decision owners visible. Any assertion without accessible proof remains pending rather than being inferred from polished language.

Define allowed categories and confidence_reason. Reject non-JSON output or unknown categories. Without model access, use labelled synthetic outputs to test the validator. Check scope, input quality, transformations, output format and approval in that order. Preserve the first attempt and justify each edit. Never hide unresolved cases simply to make the presentation look complete.

Example prompt

Training scenario: A mock classifier returns a fourth category not in the schema. Goal: JSON output contract, checks and manual fallback. Use only supplied inputs; distinguish facts, assumptions and open questions. Do not cite unseen sources. Return a table with decision, reason, evidence and status, followed by unresolved checks.

Independent exercise

Work through this scenario: A mock classifier returns a fourth category not in the schema. Deliver json output contract, checks and manual fallback. Create two versions using the same source data, check consistency, log discrepancies and name the required approving role.

Quality safeguard

A model confidence number is not proof of correctness.

Knowledge check

1. Which principle applies to the chapter “Introducing a model into the workflow”?

  1. An unverified answer is official policy.
  2. Inputs can be omitted.
  3. Deliver: JSON output contract, checks and manual fallback.
2. What should a responsible team do in a similar case?

  1. Follow the method: Define allowed categories and confidence_reason. Reject non-JSON output or unknown categories. Witho
  2. Change evidence during a comparison.
  3. Ignore missing cases for speed.
3. Which control reduces professional errors?

  1. Publish the first output immediately.
  2. Apply this safeguard: A model confidence number is not proof of correctness.
  3. Run without authorisation.

9. Human oversight and error handling

Workplace application · Intermediate to advanced

An account change request is marked urgent without evidence.

Method and workflow — TRACE method

A safe process includes stop, resume and reject paths. Human review is an explicit state transition: proposed → approved or rejected; silence is never approval. The outcome of this chapter is observable: Approval/rejection log and recovery procedure. A professional result keeps evidence, assumptions and decision owners visible. Any assertion without accessible proof remains pending rather than being inferred from polished language.

Create a protocol with reviewer, decision, timestamp and reason. Simulate an outage, ambiguous input and rejection. Verify that no case triggers external action and all are recoverable from the log. Check scope, input quality, transformations, output format and approval in that order. Preserve the first attempt and justify each edit. Never hide unresolved cases simply to make the presentation look complete.

Example prompt

Training scenario: An account change request is marked urgent without evidence. Goal: Approval/rejection log and recovery procedure. Use only supplied inputs; distinguish facts, assumptions and open questions. Do not cite unseen sources. Return a table with decision, reason, evidence and status, followed by unresolved checks.

Independent exercise

Work through this scenario: An account change request is marked urgent without evidence. Deliver approval/rejection log and recovery procedure. Create two versions using the same source data, check consistency, log discrepancies and name the required approving role.

Quality safeguard

Never let a workflow approve its own sensitive action.

Knowledge check

1. Which principle applies to the chapter “Human oversight and error handling”?

  1. Deliver: Approval/rejection log and recovery procedure.
  2. An unverified answer is official policy.
  3. Inputs can be omitted.
2. What should a responsible team do in a similar case?

  1. Ignore missing cases for speed.
  2. Follow the method: Create a protocol with reviewer, decision, timestamp and reason. Simulate an outage, ambiguous input
  3. Change evidence during a comparison.
3. Which control reduces professional errors?

  1. Run without authorisation.
  2. Publish the first output immediately.
  3. Apply this safeguard: Never let a workflow approve its own sensitive action.

10. Testing, logs and metrics

Workplace application · Intermediate to advanced

The process handles ten requests but two outputs are missing and one is duplicated.

Method and workflow — TRACE method

Reliability metrics include input coverage, errors, false positives, latency and human-review volume. Evaluating on a reference set means preserving failure cases, not just successes. The outcome of this chapter is observable: QA report with exact counts and documented discrepancies. A professional result keeps evidence, assumptions and decision owners visible. Any assertion without accessible proof remains pending rather than being inferred from polished language.

Prepare ten normal and four edge cases. Compute accuracy, coverage and human-review counts from recorded outputs. Recompute against reference labels and document stop thresholds. Check scope, input quality, transformations, output format and approval in that order. Preserve the first attempt and justify each edit. Never hide unresolved cases simply to make the presentation look complete.

Example prompt

Training scenario: The process handles ten requests but two outputs are missing and one is duplicated. Goal: QA report with exact counts and documented discrepancies. Use only supplied inputs; distinguish facts, assumptions and open questions. Do not cite unseen sources. Return a table with decision, reason, evidence and status, followed by unresolved checks.

Independent exercise

Work through this scenario: The process handles ten requests but two outputs are missing and one is duplicated. Deliver qa report with exact counts and documented discrepancies. Create two versions using the same source data, check consistency, log discrepancies and name the required approving role.

Quality safeguard

Do not claim universal reliability from a small test.

Knowledge check

1. Which principle applies to the chapter “Testing, logs and metrics”?

  1. Inputs can be omitted.
  2. Deliver: QA report with exact counts and documented discrepancies.
  3. An unverified answer is official policy.
2. What should a responsible team do in a similar case?

  1. Change evidence during a comparison.
  2. Ignore missing cases for speed.
  3. Follow the method: Prepare ten normal and four edge cases. Compute accuracy, coverage and human-review counts from reco
3. Which control reduces professional errors?

  1. Apply this safeguard: Do not claim universal reliability from a small test.
  2. Run without authorisation.
  3. Publish the first output immediately.

11. Governance, costs and privacy

Workplace application · Intermediate to advanced

A hypothetical budget rises after adding a usage-based service.

Method and workflow — TRACE method

Automations need an owner, spending cap, retention policy and stop procedure. Platform pricing changes; distinguish observed cost from hypothetical training scenarios. The outcome of this chapter is observable: Governance sheet, stop thresholds and cost assumptions. A professional result keeps evidence, assumptions and decision owners visible. Any assertion without accessible proof remains pending rather than being inferred from polished language.

Classify data sensitivity and external services. Set usage alerts and approve new connectors before activation. Review least privilege whenever the flow changes. Check scope, input quality, transformations, output format and approval in that order. Preserve the first attempt and justify each edit. Never hide unresolved cases simply to make the presentation look complete.

Example prompt

Training scenario: A hypothetical budget rises after adding a usage-based service. Goal: Governance sheet, stop thresholds and cost assumptions. Use only supplied inputs; distinguish facts, assumptions and open questions. Do not cite unseen sources. Return a table with decision, reason, evidence and status, followed by unresolved checks.

Independent exercise

Work through this scenario: A hypothetical budget rises after adding a usage-based service. Deliver governance sheet, stop thresholds and cost assumptions. Create two versions using the same source data, check consistency, log discrepancies and name the required approving role.

Quality safeguard

Never log credentials, tokens or customer data.

Knowledge check

1. Which principle applies to the chapter “Governance, costs and privacy”?

  1. An unverified answer is official policy.
  2. Inputs can be omitted.
  3. Deliver: Governance sheet, stop thresholds and cost assumptions.
2. What should a responsible team do in a similar case?

  1. Follow the method: Classify data sensitivity and external services. Set usage alerts and approve new connectors before
  2. Change evidence during a comparison.
  3. Ignore missing cases for speed.
3. Which control reduces professional errors?

  1. Publish the first output immediately.
  2. Apply this safeguard: Never log credentials, tokens or customer data.
  3. Run without authorisation.

12. Capstone: supervised automation

Workplace application · Intermediate to advanced

A fictional organisation wants to triage tickets before staff action.

Method and workflow — TRACE method

The capstone combines input contracts, triage rules, a mock API, test logs, failure cases and human sign-off. Success means accurate traceable outputs, not a flashy demonstration. The outcome of this chapter is observable: Code, results, tests, diagram, approval log and presentation. A professional result keeps evidence, assumptions and decision owners visible. Any assertion without accessible proof remains pending rather than being inferred from polished language.

Run the forty synthetic requests through the local script. Test duplicates, empty fields and unknown categories. Deliver the output CSV, exception report, simulated human decision and recovery plan. No live emails or paid API calls. Check scope, input quality, transformations, output format and approval in that order. Preserve the first attempt and justify each edit. Never hide unresolved cases simply to make the presentation look complete.

Example prompt

Training scenario: A fictional organisation wants to triage tickets before staff action. Goal: Code, results, tests, diagram, approval log and presentation. Use only supplied inputs; distinguish facts, assumptions and open questions. Do not cite unseen sources. Return a table with decision, reason, evidence and status, followed by unresolved checks.

Independent exercise

Work through this scenario: A fictional organisation wants to triage tickets before staff action. Deliver code, results, tests, diagram, approval log and presentation. Create two versions using the same source data, check consistency, log discrepancies and name the required approving role.

Quality safeguard

All irreversible actions remain disabled pending explicit authorisation.

Knowledge check

1. Which principle applies to the chapter “Capstone: supervised automation”?

  1. Deliver: Code, results, tests, diagram, approval log and presentation.
  2. An unverified answer is official policy.
  3. Inputs can be omitted.
2. What should a responsible team do in a similar case?

  1. Ignore missing cases for speed.
  2. Follow the method: Run the forty synthetic requests through the local script. Test duplicates, empty fields and unknown
  3. Change evidence during a comparison.
3. Which control reduces professional errors?

  1. Run without authorisation.
  2. Publish the first output immediately.
  3. Apply this safeguard: All irreversible actions remain disabled pending explicit authorisation.

Check your knowledge

1. Automation, assistants and agents
1. Which principle applies to the chapter “Automation, assistants and agents”?
2. What should a responsible team do in a similar case?
3. Which control reduces professional errors?
2. Architecture: triggers, tools and state
1. Which principle applies to the chapter “Architecture: triggers, tools and state”?
2. What should a responsible team do in a similar case?
3. Which control reduces professional errors?
3. Scoping goals and permissions
1. Which principle applies to the chapter “Scoping goals and permissions”?
2. What should a responsible team do in a similar case?
3. Which control reduces professional errors?
4. A first supervised workflow
1. Which principle applies to the chapter “A first supervised workflow”?
2. What should a responsible team do in a similar case?
3. Which control reduces professional errors?
5. Structuring and validating data
1. Which principle applies to the chapter “Structuring and validating data”?
2. What should a responsible team do in a similar case?
3. Which control reduces professional errors?
6. No-code platforms: n8n, Make and Zapier
1. Which principle applies to the chapter “No-code platforms: n8n, Make and Zapier”?
2. What should a responsible team do in a similar case?
3. Which control reduces professional errors?
7. API connections without real secrets
1. Which principle applies to the chapter “API connections without real secrets”?
2. What should a responsible team do in a similar case?
3. Which control reduces professional errors?
8. Introducing a model into the workflow
1. Which principle applies to the chapter “Introducing a model into the workflow”?
2. What should a responsible team do in a similar case?
3. Which control reduces professional errors?
9. Human oversight and error handling
1. Which principle applies to the chapter “Human oversight and error handling”?
2. What should a responsible team do in a similar case?
3. Which control reduces professional errors?
10. Testing, logs and metrics
1. Which principle applies to the chapter “Testing, logs and metrics”?
2. What should a responsible team do in a similar case?
3. Which control reduces professional errors?
11. Governance, costs and privacy
1. Which principle applies to the chapter “Governance, costs and privacy”?
2. What should a responsible team do in a similar case?
3. Which control reduces professional errors?
12. Capstone: supervised automation
1. Which principle applies to the chapter “Capstone: supervised automation”?
2. What should a responsible team do in a similar case?
3. Which control reduces professional errors?

Glossary for this page

The key terms used in this content, each with a short definition.

API
An interface that lets two programs exchange data and commands in a standard way.
automation
Making a repetitive task run by itself according to a rule: when this happens, do that.
workflow
The sequence of steps a task follows from start to finish, across people or apps.
no-code
Building sites, apps and automations with visual interfaces instead of writing code.
token
A small unit of text (a word or word fragment) processed by the model; it measures text length and conversation limits.
CSV
A plain text file with delimiter-separated values; the most common format for moving data between programs.

Course files

Choose a language, then the format you need.

العربية

FileFormatSizeDownload
الدليل المهني الشامل PDF 746 KB Download — الدليل المهني الشامل (PDF)
كراسة التمارين PDF 491 KB Download — كراسة التمارين (PDF)
كتاب التصحيح المشروح PDF 518 KB Download — كتاب التصحيح المشروح (PDF)
المشروع النهائي التطبيقي PDF 469 KB Download — المشروع النهائي التطبيقي (PDF)
بطاقات منهجية PDF 479 KB Download — بطاقات منهجية (PDF)
الأمان والخصوصية PDF 475 KB Download — الأمان والخصوصية (PDF)
دليل البدء PDF 472 KB Download — دليل البدء (PDF)
مكتبة 36 تعليمة مشروحة CSV 27 KB Download — مكتبة 36 تعليمة مشروحة (CSV)
40 طلباً اصطناعياً مع التصنيف المرجعي CSV 2 KB Download — 40 طلباً اصطناعياً مع التصنيف المرجعي (CSV)
سجل ادعاءات للتحقق CSV 4 KB Download — سجل ادعاءات للتحقق (CSV)
شبكة تقييم المهارات CSV 2 KB Download — شبكة تقييم المهارات (CSV)
مصنف تقييم مهني XLSX XLSX 6 KB Download — مصنف تقييم مهني XLSX (XLSX)
سير عمل تخطيطي JSON JSON 1 KB Download — سير عمل تخطيطي JSON (JSON)
فحص ما قبل النشر TXT 2 KB Download — فحص ما قبل النشر (TXT)
كراسة التمارين TXT 11 KB Download — كراسة التمارين (TXT)
حقيبة تطبيقية: بيانات وقوالب دون ملفات PDF ZIP 17 KB Download — حقيبة تطبيقية: بيانات وقوالب دون ملفات PDF (ZIP)
Offline deterministic triage demo Shared Python 2 KB Download — Offline deterministic triage demo (Python)
Reference output for 40 requests Shared CSV 3 KB Download — Reference output for 40 requests (CSV)
n8n sample not tested in native app Shared JSON 1 KB Download — n8n sample not tested in native app (JSON)

Français

FileFormatSizeDownload
Manuel professionnel PDF 740 KB Download — Manuel professionnel (PDF)
Cahier d’exercices PDF 481 KB Download — Cahier d’exercices (PDF)
Corrigés commentés PDF 512 KB Download — Corrigés commentés (PDF)
Projet final guidé PDF 460 KB Download — Projet final guidé (PDF)
Fiches méthodes PDF 471 KB Download — Fiches méthodes (PDF)
Sécurité et confidentialité PDF 468 KB Download — Sécurité et confidentialité (PDF)
Guide de prise en main PDF 464 KB Download — Guide de prise en main (PDF)
Bibliothèque commentée de 36 consignes CSV 24 KB Download — Bibliothèque commentée de 36 consignes (CSV)
40 demandes synthétiques avec catégories de référence CSV 2 KB Download — 40 demandes synthétiques avec catégories de référence (CSV)
Registre des affirmations à vérifier CSV 2 KB Download — Registre des affirmations à vérifier (CSV)
Grille de contrôle des compétences CSV 2 KB Download — Grille de contrôle des compétences (CSV)
Matrice professionnelle XLSX XLSX 6 KB Download — Matrice professionnelle XLSX (XLSX)
Workflow conceptuel supervisé JSON JSON 1 KB Download — Workflow conceptuel supervisé JSON (JSON)
Validation avant publication TXT 2 KB Download — Validation avant publication (TXT)
Cahier d’exercices TXT 10 KB Download — Cahier d’exercices (TXT)
Kit pratique : données, matrices et modèles (sans PDF) ZIP 17 KB Download — Kit pratique : données, matrices et modèles (sans PDF) (ZIP)
Offline deterministic triage demo Shared Python 2 KB Download — Offline deterministic triage demo (Python)
Reference output for 40 requests Shared CSV 3 KB Download — Reference output for 40 requests (CSV)
n8n sample not tested in native app Shared JSON 1 KB Download — n8n sample not tested in native app (JSON)

English

FileFormatSizeDownload
Professional course manual PDF 733 KB Download — Professional course manual (PDF)
Practice workbook PDF 479 KB Download — Practice workbook (PDF)
Worked answer book PDF 508 KB Download — Worked answer book (PDF)
Guided capstone project PDF 459 KB Download — Guided capstone project (PDF)
Method cards PDF 468 KB Download — Method cards (PDF)
Security and privacy PDF 465 KB Download — Security and privacy (PDF)
Getting-started guide PDF 462 KB Download — Getting-started guide (PDF)
36 annotated practice prompts CSV 19 KB Download — 36 annotated practice prompts (CSV)
40 synthetic requests with reference labels CSV 2 KB Download — 40 synthetic requests with reference labels (CSV)
Claim-evidence register CSV 2 KB Download — Claim-evidence register (CSV)
Competency check rubric CSV 1 KB Download — Competency check rubric (CSV)
Professional workbook XLSX XLSX 5 KB Download — Professional workbook XLSX (XLSX)
Supervised conceptual JSON flow JSON 1 KB Download — Supervised conceptual JSON flow (JSON)
Prepublication validation TXT 1 KB Download — Prepublication validation (TXT)
Practice workbook TXT 8 KB Download — Practice workbook (TXT)
Practical kit: data, grids, templates (no PDF) ZIP 15 KB Download — Practical kit: data, grids, templates (no PDF) (ZIP)
Offline deterministic triage demo Shared Python 2 KB Download — Offline deterministic triage demo (Python)
Reference output for 40 requests Shared CSV 3 KB Download — Reference output for 40 requests (CSV)
n8n sample not tested in native app Shared JSON 1 KB Download — n8n sample not tested in native app (JSON)