№ 01 · Field Kits for AI agents
Your AI never
says what it
didn’t check.
It sounds sure, so nobody asks what it skipped. You already have the AI. Conclusion MP gives it a professional method.
Example: Ask “This CPA jumped 40% overnight. Is it real?” and the installed skill rules out paused campaigns, budget caps and tracking breaks before it names a cause. See it work
- Field Kits
- 21
- skills
- 126
- subagents
- 21
- commands
- 22
21 Field Kits · 10 manuals · one checkout · Instant download · 14‑day money back · No coding · Runs in Claude Code and claude.ai · VAT handled at checkout
The Complete Collection
How it works
One minute to install. Every week it pays back.
A kit is a folder of instructions your agent reads before it answers. You install it once; from then on the question is enough, because the method is already there.
01 · Pick a kit
The Reporting Engine
6 skills · 1 subagent · 1 command
One job your team repeats every week, written down as a method.
02 · Install it
.claude/skills/ anomaly-triage/ SKILL.mdCopy the folder into Claude Code, or upload the zip to claude.ai. No coding.
03 · Ask as usual
“This CPA jumped 40% overnight. Is it real?”
No prompt to write. The agent recognises the job and loads the skill.
04 · Same answer, every time
- Severity
- S3 · watch
- Likely cause
- Delayed attribution
- Evidence
- Clicks held; the last two days are still inside the attribution window
- Alert client
- No
- Next check
- Thursday, re-export the trailing 7 days
Example row from the triage log
This is the skill we give away. Try one skill free
The objection
Not a prompt library.
A prompt tells an agent what you want, once. A skill states when the workflow applies, what it requires before it starts, what process it follows in order, what it refuses to invent, and what the finished deliverable contains. One is an answer. The other is a method that survives the person who wrote it.
Ask your agent
“This CPA jumped 40% overnight. Is it real?”
Without the kit
- A general answer, assembled from your sentence and whatever is still in the context window.
- A cause gets guessed. Seasonality, the creative, the algorithm, pick one.
- Nothing declares what it will not invent, so nothing stops it inventing.
With The Reporting Engine
anomaly-triage
- 01Build the baseline
- 02Flag deviations
- 03Run the non-cause list
- 04Grade severity
- 05Write evidence
- 06Decide the alert
- 07Re-check
Same method next week · Same method for the next analyst
Run it on your own numbers first. Try one skill free
From the people using it
In their words.
★★★★★
“The distinction between observation, explanation and decision is especially useful. It gives the report a clear structure without forcing a story the numbers don’t support.”
Marketing Analytics LeadUsed The Reporting Engine★★★★★
“The decision frameworks were the most useful part for me. They gave me a clearer way to question priorities and explain the trade-offs behind a plan.”
Founder & CEORead Think Like a CEO★★★★★
“I appreciated how it connects marketing decisions to the wider business. The sections on measurement gave me useful questions to bring into our reporting meetings.”
Head of MarketingRead Think Like a CMO★★★★★
“The decision memo structure makes you spell out the options and trade-offs before writing a recommendation. That’s the part I keep coming back to.”
Chief of StaffUsed The Executive Desk★★★★★
“The examples helped me separate what the data shows from what we’re assuming. I found that more useful than another list of metrics to track.”
Director of Product MarketingRead The Measurement Manual★★★★★
“I liked the emphasis on defining the task and reviewing the output. It gave me a more deliberate way to use AI in work that still needs my judgment.”
Head of ProductRead The Augmented Practitioner
№ 02 · Two ways in
Same method, two formats
The working side
Field Kits
Installable agent skills, subagents and slash commands your AI agent follows every time. Every kit ships six skills, a read-only subagent and a slash command.
- kits
- 21
- skills
- 126
- subagents
- 21
The reading side
Field Manuals
The argument behind each kit, as a PDF you read in an evening. Sixteen to thirty-six pages each, every figure sourced.
- manuals
- 11
- pages
- 303
№ 03 · Skills, subagents, commandsNew
Detailed instructions your AI agent follows every time.
A Field Kit is six installable agent skills, a read-only subagent and a slash command, built from the manual it pairs with. Each skill says when it applies, what it needs, what it refuses and what comes back.
All twenty-one kits · $12 each →Marketing
eight kits
Demand, brand, search, the editorial desk and the pitch that wins the account.
AI
four kits
Build the tools, buy the tools, and decide what the organization does about them.
Leadership
four kits
The operating rhythm of a leader, and where the money actually goes.
Analytics
three kits
Numbers you can defend, and measurement decided before the thing gets built.
Retail Media
two kits
Both sides of the counter: the brand buying it, the retailer selling it.
Pick a desk
One discipline, one checkout.
four kits
The Agency Desk
Four Field Kits for the people who sell the work
two manuals · four kits
The AI Shelf
Two manuals and four Field Kits for the organization that has decided
five kits
The Growth Desk
Five Field Kits for the in-house marketer
three manuals · four kits
The Leadership Shelf
Three manuals and four Field Kits for the person who owns the outcome
one manual · three kits
The Retail Shelf
The manual and three Field Kits for both sides of the counter
The Complete Operation
Ten manuals, every kit, one acquisition.
The whole library and the whole toolkit: ten field manuals and twenty-one Field Kits of installable agent skills, subagents and commands. 126 skills, $99 instead of $312.
Bought separately
- 21 Field Kits · 126 skills, 21 subagents, 22 commands
- $252
- 10 field manuals · PDF
- $59.90
- Together
- $312
- As The Complete Operation, one checkout, thirty-one downloads
- $99
Instant download · Personal licence · Revisions included · No subscription
№ 04 · The Manuals
The manuals behind the kits.
Every kit is built from a field manual that makes the argument and names the sources. Read the manual to know why the tool refuses what it refuses. Sixteen to thirty-six pages each, one idea at a time.
All eleven manuals · from $5.99 →The Standard
A model without a method is a confident intern. A method without a model is an expensive presentation.
Every manual and every kit we publish is built on that principle: no filler padded to justify a price, no skill that invents a number, no output you would have to redo before you could use it. One subject, one standard, installed in a minute.
№ 05 · Disciplines
Five subjects. One standard of rigour.
Straight answers
Questions before a first order.
01What is the difference between a manual and a kit?
A manual is the argument, as a PDF you read. A Field Kit is that method as installable agent skills your AI agent runs.
02Do I need Claude Code?
No. Every skill uploads to claude.ai as a zip. Subagents and slash commands are the Claude Code extras.
03I am not a programmer.
Install is copying folders into a .claude directory. The Field Guide in each kit assumes nothing beyond that.
04Do the tools invent numbers?
Never. Every figure is supplied by you, confirmed by you, or labeled as an assumption in the output.
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.
Begin
Install the method.
One manual, one evening. Or the whole operation installed in your AI agent by tomorrow morning.