npx skills add ...
npx skills add samber/dev-event-organizer-skills --skill event-vip-management
Receive named guests who are neither speaking nor sponsoring but whose presence creates escort, protocol, security or discretion obligations the general attendee flow cannot absorb - public officials, dignitaries, major funders, executives with no booth duty. Covers the qualification test (obligation, never status), the identification roster, escort depth, and a visibility posture that meets the obligation without building a status tier attendees can read. Use whenever the user mentions receiving a minister, mayor, funder or visiting executive, VIP badges or reserved seating, escorting a dignitary, or coordinating with a guest's own security detail - even if they never say "VIP". Do NOT use for speakers - use samber/dev-event-organizer-skills@event-speaker-experience instead.
npx skills add samber/dev-event-organizer-skills --skill event-vip-management
You receive a named guest whose visit creates an obligation your general attendee flow cannot absorb, and you do it without building a tier the room can see. Those are two jobs, and the second one is the hard one.
Two illustrative shapes:
Both are yours. Neither is a reason to print a different badge.
The qualifying fact is an obligation, never a title. A guest qualifies when their presence imposes something the door, the queue and the help desk cannot handle as they stand: an escort, a protocol requirement, a security constraint, a discretion request. Prominence alone qualifies nobody, and the most common correct outcome of this skill's interview is that nobody is left in scope.
Route these out at the interview, before anything else. Some of these boundaries are agreed from both sides, others are written from here only.
samber/dev-event-organizer-skills@event-speaker-experience takes every speaker from acceptance to thank-you, however senior, and its speaker-care menu already carries a rung that books travel and lodging rather than reimbursing, promoted for the international or visa-dependent traveller. A second owner for one relationship is how a commitment gets dropped between two runbooks.samber/dev-event-organizer-skills@event-sponsor-fulfillment delivers every promised perk against the agreement and gates each on payment. A sponsor's executive on the clock at their booth is theirs. You start only where a guest's presence is covered by no agreement at all: a sponsor CEO attending purely as a guest, with no booth duty. That edge is written from this side only; fulfillment never mentions it.samber/dev-event-organizer-skills@event-press-relations claims accreditation, embargoes and on-site press handling, and states that boundary from its own side as well. You own only a guest's own exposure to press: declining a photograph on their behalf, keeping a departure time off the schedule.samber/dev-event-organizer-skills@event-attendee-experience designs the door, the queue and the help path, and names you from its side: "a VIP is a separate protocol track, never the top rung of this journey." You never upgrade an attendee into this skill; you route a guest out of it.You add only the guest-specific layer on top of the shared floor samber/dev-event-organizer-skills@event-hospitality sets. That skill ships and decides:
It names this boundary from its own side, delivering the floor and never re-deciding "a speakers' dinner, a volunteer meal or a VIP table". A guest's table is therefore yours; the food on it is not. Route the user there when the real question is a hospitality question, and do not restate one of its four decisions here.
All four menus below are this skill's own construction. Their rungs, axis orderings, defaults, promotion conditions, and delete rules are argued here, not drawn from a published standard or measured against event data.
Every ranking below is a default, not a law: it shifts with context and with who executes it. After the interview, re-rank all four menus against what you know. Each of these overturns a default rung, and the list is not closed:
Ask one question at a time, multiple-choice where possible. Questions 5-8 exist because the menus below diverge sharply on visibility, time-to-effect and effort - those defaults cannot be picked for the user.
Who owns the event, and what it is for, flips a whole menu. Sibling skills in this collection argue the same split.
samber/dev-event-organizer-skills@event-cultural-identity set. The obligation still has to be met - it just has to be met invisibly.Say which pole a recommendation assumes whenever they differ. Menu B is the only one of the four where they genuinely diverge; A, C and D hold their order across both.
Run the interview. Route out speakers, sponsor personnel and press first. If nobody remains in scope, say so and stop - that is a correct result, not a failed one.
Apply the qualification test to whoever remains: obligation, not status. The test, its four qualifying obligation types, and the worked pass/fail pairs are in references/guest-qualification-and-roster.md.
Set the identification breadth (Menu D), then build the roster: its fields, its named holder, who else sees it, and its deletion date - decided now, not after the event. Same reference.
Set the program posture (Menu A).
Set the visibility posture (Menu B) against the pole (Q2) and against what the guest actually said (Q5). Where there is still time, ask rather than decide; a guest who has not been asked is a guess, not a preference.
Set the escort depth (Menu C) and write the greeter brief. Template, the what-not-to-say list, and the discretion mechanics for each Menu B rung are in references/discretion-and-escort-mechanics.md.
Where a guest arrives with their own detail or advance team, run the liaison intake in that same reference:
samber/dev-event-organizer-skills@event-code-of-conduct names contractors and press in its scope clause; somebody else's security staff is the same shape, and the clause wording is theirs, not yours.
Write down the two decisions the day will test, with a name against each: who decides on the spot when a photograph is requested, and who says no on the guest's behalf. Name someone in advance, or whoever is standing there decides both under time pressure.
Hand off what is not yours:
samber/dev-event-organizer-skills@event-schedule-design and samber/dev-event-organizer-skills@event-run-of-show.samber/dev-event-organizer-skills@event-attendee-experience.samber/dev-event-organizer-skills@event-accessibility-inclusion, because a guest with an access need is an access question first and a guest question second.samber/dev-event-organizer-skills@event-risk-management.Present section by section - qualification, roster, posture, visibility, escort - and get explicit approval per section before finalizing.
If your harness has persistent memory, record per edition:
Next edition starts from that instead of re-deriving a posture for the same guest.
How much apparatus exists at all, before any specific guest arrives.
formal named program with dedicated staff > defined criteria + protocol runbook > ad hoc invisible accommodation > no postureformal named program > defined criteria + runbook > ad hoc invisible accommodation > no postureformal named program > defined criteria + runbook == ad hoc invisible accommodation > no posturead hoc invisible accommodation > defined criteria + runbook > formal named program > no postureValue and effort run in the same order on this menu. No rung offers more value at less effort, so zero dominance relations exist and the strict-dominance check cannot find anything. This menu is clean by construction, not by care, and that is not a pass: the efficiency line rests entirely on each rung's own argument below.
Compliance cost is the one axis that departs from that order, and its == is argued. The runbook triggers a data-handling review that ad hoc skips; ad hoc spends the reversibility the runbook keeps, because an undocumented list of names in one organizer's messages has no stated holder, no deletion date, and no way to confirm it is gone. Equal exposure, opposite halves.
Ad hoc invisible accommodation - the default. One organizer quietly handles whatever a specific guest needs; nothing is written, nothing is announced, nothing is repeatable. It wins on ratio because the common case is one or two guests a year and the whole obligation is met by one person paying attention. Move up one rung when Q7 says the relationship recurs and Q9 puts more than a handful of names on the list - past that, the arrangement lives in one person's head and leaves with them.
Defined criteria + protocol runbook - documented qualifying triggers, a named owner, a repeatable sequence, and a roster with a deletion date. The promotion target for a recurring institutional relationship.
Formal named program with dedicated staff - the starved rung: top of value, top of effort, top of compliance cost, so efficiency never picks it. Promotion condition: Q4 says a guest arrives with a protective detail or advance team and Q7 says this recurs - at that point you are the standing counterpart to somebody else's standing process, and improvising against it each edition is the actual risk.
No posture - every guest absorbed into the general, speaker or sponsor flow, unchanged. It sits last on efficiency for the same reason samber/dev-event-organizer-skills@event-cultural-identity puts "no deliberate ritual" last: near-zero effort buying nothing on the axis being measured is a ratio of zero, and by the time this skill is running, a qualifying guest already exists.
Promote no posture to first outright when Q3 finds status without obligation: the guest is prominent, arrives alone, brings no detail, states no requirement. Every rung above it then manufactures a tier for nothing. That promotion is not a courtesy, and at the community pole it is the most common right answer.
Deleted when Q2 says a flat, explicitly anti-hierarchy register and Q4 finds no external obligation: the formal named program. Delete it from this menu and from the axis lines above. A standing VIP program at an event whose stated identity refuses status is not an expensive option to defer - it is a contradiction that will be read as one, and parked at the bottom it comes back next planning cycle as a proposal.
What the handling shows. This is the menu carrying the cultural tension, and the only one whose value ordering flips by pole.
public marker + recognition moment > visible marker > staff-only signal > invisible handlingstaff-only signal > invisible handling > visible marker > public marker + recognition momentpublic marker + recognition moment > visible marker > staff-only signal > invisible handlingstaff-only signal > invisible handling > visible marker > public marker + recognition momentvisible marker > staff-only signal > public marker + recognition moment > invisible handlingDominance check, community pole - clean by care. Value and effort run opposite, so five dominance relations exist and each one could have been violated:
The community-pole efficiency line honours all five. Only staff-only signal against invisible handling is unconstrained - value favours the first, effort the second - and it is decided by ratio below, not by dominance.
Dominance check, vendor/corporate pole - clean by construction. At this pole value and effort run in the same order, so there are no dominance relations at all and the check finds nothing. The pole flip converts this menu from checkable to unfalsifiable, and the vendor-pole efficiency line rests on its own argument alone.
No compliance-cost axis is printed here, and the skip is argued: rung for rung it would reproduce the effort ordering exactly, adding a line and no information. What it would have carried is the reversibility asymmetry, and that belongs in the rungs themselves: a badge colour can be swapped between sessions, and a stage announcement and a published photograph cannot be unsaid.
Interaction with Menu D, and it runs one way: widening the identification list pushes this menu toward its costlier, more visible rungs. Four obligation-carrying guests can be held in one organizer's head and briefed by one message. Forty cannot, so they need a marker somebody can read at a glance, which is a visible tier arrived at by logistics rather than by decision. If this menu's outcome is a visible marker and nobody ever chose one, check Menu D first.
How much of an organizer's day a guest gets, and at which moments.
escort + protocol/security liaison > dedicated escort for the visit > named greeter for arrival only > printed contact cardescort + liaison > dedicated escort > named greeter > printed contact cardnamed greeter for arrival only > printed contact card > dedicated escort for the visit > escort + liaisonValue and effort run in the same order here too: zero dominance relations, clean by construction, not a pass. The efficiency line rests on each rung's argument.
No compliance-cost axis, and the skip is argued: exactly one rung carries a review - the liaison rung - and that review is not yours to run. It belongs to the guest's own institution, which arrives with it already done. A four-rung axis with one non-zero entry is not an ordering.
Who goes on the list at all.
broad notability list > obligation + relationship list > obligation-only list > no defined listbroad notability list > obligation + relationship list > obligation-only list > no defined listbroad notability list > obligation + relationship list > obligation-only list == no defined listobligation-only list > obligation + relationship list > no defined list > broad notability listValue and effort run in the same order again: zero dominance relations, clean by construction, not a pass. Three of this skill's four menus are in that position; only Menu B's community pole is genuinely checkable. Say so to the user if they ask why the orderings look mechanical: they are not derived, they are argued.
The compliance axis is the one that differentiates, and its == is argued. No defined list creates no artifact to review, and it leaves an organizer improvising at the door about which named individuals get different treatment, with no criteria and no record of why. An unreviewable decision is not a low-exposure one; it is an unauditable one.
The obligation-only list creates a small named-person record that triggers exactly one retention decision and is defensible line by line, because every entry is there for a reason the guest's own institution stated. Equal exposure, opposite shapes.
These eight are this skill's own construction, not an incident record.
All four signals are this skill's own construction; none is a benchmark, and no published figure exists to calibrate any of them against. Fix each target before the event, never after the data arrives.
Expected output: a guest brief containing:
Presented section by section for approval. Where nobody qualifies, the output is that finding and nothing else.