The Coexistence Desk
Skills · Subagent · Command · PDF
.claude/the-coexistence-desk/zip
skills/06
├ agent-action-inventory
├ ai-programme-charter
├ decision-friction-protocol
├ governance-review
├ role-redesign-brief
├ workflow-selector
agents/01
├ ai-governance-auditor
commands/01
├ /inventory
FIELD-GUIDE.pdf9pp
English
One payment · No account · No subscription
- ✓ Download link in your inbox in seconds
- ✓ Secure card checkout · VAT handled
- ✓ 14-day money back, no questions
- ✓ Every future revision included
- ✓ Runs in Claude Code and claude.ai
Built for Claude first, our recommended agent. Skills follow the open Agent Skills standard (SKILL.md), so they also run in any agent that reads it; subagents and slash commands are Claude Code formats.
The Coexistence Desk
The twelve organizational decisions, as tools
six skills·one subagent·one command·9-page guide
The twelve organizational decisions behind an AI programme, produced as documents: a charter with a number in it, a shortlist of workflows, an inventory of what each agent may do.
Executive summary
- The problem
- Most AI programmes report adoption, not value: an ambition with no number attached, a spread of disconnected pilots, and agents taking external actions nobody has written down. It stays invisible until someone asks what an agent was allowed to do, and no document answers.
- What it does
- ai-programme-charter marks the target missing when nobody set one, workflow-selector cuts the shortlist to the few worth reinventing, agent-action-inventory gives every external action a risk class, and the governance-auditor subagent lists the gaps.
- Why take it
- You are answerable for what your organization's AI actually does. These are the documents that show it, before a regulator or a board asks for them.
For executives and programme owners answerable for what an organization's AI actually does.
In the first week you will
- 01Write the one-page programme charter with ai-programme-charter and see the target marked missing when nobody attached a figure
- 02Run /inventory on an agent before it goes live and get every external action with its risk class and autonomy threshold
- 03Send your policies and pilot reports to the ai-governance-auditor subagent and get the gap list against the twelve decisions
Contents of the Kit
- 01Six installable agent skills: ai-programme-charter, workflow-selector, agent-action-inventory, role-redesign-brief, governance-review, decision-friction-protocol
- 02The ai-governance-auditor subagent: reads your policies, agent configurations and pilot reports, and returns the gap list against the twelve decisions. Read-only
- 03An /inventory slash command that opens the external-action inventory for an agent from one sentence
- 04The agent inventory in the shape European law asks for: external actions, data flows, connected systems, affected persons, a risk class and an autonomy threshold per action
- 05Blank templates for the agent action inventory and the decision friction log
- 06A 9-page Field Guide on the order the decisions pay off in, and plugin manifest included
One skill, as it ships
SKILL.md
ai-programme-charter
- Use when
- Use when a leadership team is starting or restarting an AI programme, when a board asks what the programme is for, when an AI budget is being approved without a target, or when someone asks "what are we actually trying to achieve with AI". Returns a draft with the target marked missing if no number is supplied.
- Requires
- The ambition, in the leadership team's words · The number · The owner · The ownership model · The scoreboard
- Will not
- No target, no charter · Usage is never value · The open items list is the last section and is never cut to make room
- Returns
- AI programme charter: {organisation} · {horizon}
One of six skills in the kit · All of them install the same way
The contents, in full · six skills, one subagent, one command
- 01ai-programme-charterDecisions one, two and four on one page: the multiyear ambition with a cost or revenue figure attached, the owner and the shared business-and-IT ownership model with decision rights, and the scoreboard that reports value rather than adoption, function by function.
- 02workflow-selectorDecision three. A list of candidate processes in, the few to reinvent end to end out, scored against criteria you weight, with the disconnected initiatives to stop named explicitly.
- 03agent-action-inventoryDecisions five to seven. For one agent: every external action, data flow, connected system and affected person, each action assigned a risk class and a human-in-the-loop or out-of-the-loop threshold, plus the context it needs (data, process, tools, boundaries) before it goes live.
- 04role-redesign-briefDecisions eight and nine. A role or team in, the redesign out: what moves to the model, what stays, the new roles it creates, the learning time, and how the entry-level pipeline is replaced when agents absorb the work that used to train juniors.
- 05governance-reviewDecisions ten and eleven. The standing review pack: what worked, what is remeasured, what changed in the models and agents since last time, the signals that trigger reallocation, and the assumptions due for revisiting.
- 06decision-friction-protocolDecision twelve. For a chosen decision: the human estimate recorded before the model is consulted, structured disconfirmation, both positions logged, and the error attributed afterwards. The cheapest tool in the kit and the one the manual says is most likely to be cut.
- 07ai-governance-auditor (subagent)Reads policies, agent definitions, vendor documents and pilot reports, and reports the gaps against the twelve decisions with the source of each finding. It reads; it changes nothing.
- 08/inventoryOpens the external-action inventory for one agent from a sentence about what it does and what it can reach. It collapses nothing into an "other" row and marks no action as go until you have confirmed its threshold and named where its log lives.
The Age of Coexistence with AI ends with a playbook: twelve organizational decisions, sequenced in the order they pay off, assembled from the research on the five percent of companies that capture value and the sixty percent that do not. The tools hold the manual’s position. A workflow shortlist longer than the manual’s “few” is cut and the cut is shown, a role is rewritten around the model with the entry-level pipeline replaced, and a handful of decisions record the human judgment before the machine is consulted. None of the tools invent a figure: every percentage, cost and target is supplied by you or labeled as an assumption.
Pairs with
The Age of Coexistence with AI makes the argument and names the sources. This kit produces the documents the argument asks for, in the order the manual says they pay off.
Before you buy
01Do I need Claude Code?
No. Every skill uploads to claude.ai as a zip. Subagents and slash commands are the Claude Code extras.
02I am not a programmer.
Install is copying folders into a .claude directory. The Field Guide in each kit assumes nothing beyond that.
03Will it work in my agent?
Skills follow the open Agent Skills standard (SKILL.md), so any agent that reads it can run them. Built for Claude first, our recommended agent.
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.
Ready when you are
The Coexistence Desk, $12.
Instant download, one payment, every future revision included. Or take The AI Shelf, six items for $39, and this kit works out at $6.50.