The Augmented Practitioner

Five agent skills, twelve connectors and the manual that runs them

25 pages·PDF·Read in an evening

Everyone uses AI and few use it well; this is the method that survives client work, with five installable skills that carry a campaign from brief to monthly report.

Executive summary

The problem: Re-briefing a chatbot from scratch every morning feels like using AI, but the quality then depends on who typed the prompt that day, and the wrong answers arrive with the same confidence as the right ones. Nobody can see where the tool stopped being competent.
What it does: Five installable skills that carry a campaign from brief intake to the monthly report, a guide to twelve connectors, and a manual on the prompt as a brief, the Delegation Ladder and where verification belongs.
Why take it: Client work does not forgive a confident wrong number. This is the method that lets you delegate the production and still stand behind the result.

For digital marketers and agency practitioners who want a working setup rather than a better chatbot.

In the first week you will

01Install the five skills and run a real client brief through intake, media plan and ad copy in one session
02Place every repeating job on the Delegation Ladder and put the verification checkpoint where the manual says it belongs
03Connect the three sources closest to your daily work and run the question that proves a connection is live

Contents of the Dossier

01Five installable agent skills: brief intake, media plan, ad copy, monthly report, SEO audit, chained through the life of a campaign
02A semi-agent setup: the skills trigger on the job at hand, the connectors hand the model your analytics, CRM, database and browser
03Step-by-step implementation: install in Claude Code or on claude.ai in minutes, with a README and one example call per skill
04A connector setup guide: 12 connectors and MCP servers mapped to the jobs, with a suggested order and a verification routine
0525-page field manual on the method: the prompt as a brief, four worked workflows, the Delegation Ladder, the discipline
06An insight extraction log citing every source behind every claim

The contents, in full · five skills, seven parts

01campaign-brief-intakeRaw client material in (email thread, document, meeting notes), a normalized brief out, plus the gap list of questions the client still owes you and a ready-to-send follow-up email.
02media-plan-builderA bulk budget in, channel allocation, flighting and estimated deliverables out, with every benchmark supplied or confirmed by you and every assumption logged. It never invents a price.
03ad-copy-studioAn offer and a landing page in, Meta copy variants and a full Google RSA set out, character limits enforced on every line, pinning strategy suggested.
04monthly-performance-reportRaw platform exports in, a defensible monthly report out: correct aggregation first, narrative second, with a warning on every metric that does not compare across platforms.
05seo-site-auditA URL in, a weighted SEO scorecard out: critical issues, quick wins with effort estimates, and fixes concrete enough to hand to a developer.
06Part I: The Proficiency GapThe evidence that using AI and using it well are different professions.
07Part II: The Prompt as a BriefThe five-part anatomy of a working brief, decomposition, and iteration as management.
08Part III: The WorkflowFour repeating jobs worked end to end, with the checkpoints where verification actually belongs.
09Part IV: The Delegation LadderFive rungs from autocomplete to autonomous agents, and the failure modes waiting at the top.
10Part V: The Connected MachineThe twelve connections, mapped to the jobs, with the one rule that keeps connection safe.
11Part VI: The Skill LibraryWhy a skill beats re-prompting, the five in the box, and the method for building your own.
12Part VII: The DisciplineThe refusal list, procedural verification, and what never enters a model.

This is the one product in the library that is not primarily reading material. It is a semi-agent setup: each repeating job packaged as a skill the model loads on demand, your data sources connected once, and you the editor while the setup does the production work.

The manual is why it holds up under client work. The same tool that lifted consultants’ output quality by over 40% inside its competence made them 19 percentage points more likely to be wrong outside it. The difference was never the tool. It was the method, and method is learnable.

Pairs with

The Practitioner Desk is the discipline of Parts III and IV as installable tools: the Delegation Ladder as a placement test, the prompt library entry, the verification protocol, the session log, and a read-only output verifier. The Age of Coexistence with AI is the organizational half of the same argument. The Complete Dossier carries both alongside every other manual in the library.

Before you buy

01What format is it?

A PDF, typeset for screen and for print. No app, no reader account, no expiry.

02How long is a manual?

Short on purpose. One subject, one evening, nothing padded to justify a price.

03Who is it written for?

Operators who have to decide something on Monday: marketing and analytics leads, founders, agency practitioners.

04Can I buy them together?

Yes. The Complete Dossier is ten manuals in one checkout, well under the price of the parts.

05How does it arrive?

Instantly. Pay by card and the download link is in your inbox within seconds, and on the confirmation screen.

06Is this a subscription?

No. One payment, lifetime access to your download, every future revision of what you bought included.

07What if it is not useful to me?

Write to info@conclusionmp.com within 14 days of purchase and you get a full refund. No questions, no forms.

Ready when you are

The Augmented Practitioner, $5.99.

Instant download, one payment, every future revision included. Or take The Complete Dossier, ten items for $29, and this manual works out at $2.90.

Get the manual · $5.99 The Complete Dossier · $29