Murder Mystery Architect AI Starter — Host Director for Invitations, Pacing & the Sealed Reveal

From AGENTORY — ready-to-use AI Task Kits. Drop in. Power up.

One agent from the Murder Mystery Architect kit — complete, nothing cut down. Works with the AI you already have: upload the file, tell it "Start", answer a few questions, and get to work.

One specialist murder mystery agent — running inside the AI you already use.

No new app. No subscription. This agent works with any AI chat — ChatGPT, Gemini, Claude, Copilot, whichever you have (free versions work).

The Host Director (Event Producer) turns a checked case into the party itself (8 calibration questions): date and venue, food plan, your guest list with your casting notes, invitation channel and RSVP deadline, your production tools, awards, ambience, and end time. It casts your guests using only the notes you supply about them — nothing inferred, nothing looked up — drafts the invitations (each carrying only that guest's public introduction, never a fact about any other character), writes the run-of-show with dinner pacing and your host scripts for the public beats, and runs the accusation and voting round and the awards: Best Detective, Best Costume, Best Actor. It also owns production of the event package — per-character booklets (each one a full player guide: detailed story briefing, dialogue propositions for every act, complete costume guide with its image or its labelled image prompt), the host Event Guide, printable clue and evidence cards, invitations, and the sealed reveal pack — and production is gated: it will not assemble the package until the Clue Master's checks are recorded as passing for the current version of the case and every guest's player guide has passed the Character Writer's completeness check for that same version. What it assembles depends honestly on your mode: actual verified PDF files where your Docs or Canva is genuinely connected, or print-ready content with layout specs and assembly steps in a plain chat — the images themselves generated in the chat wherever the session can generate them, an image prompt only as the labelled fallback where it cannot — labelled as such, never as a finished illustrated PDF. The reveal stays sealed throughout: ballots are collected first, then the reveal — a read-aloud script, always as text, performed cold in front of everyone — is opened blind; a generated audio version appears only where your platform genuinely has audio capability in that session, otherwise you get the script plus phone-recording instructions. And if a guest drops out, the dropout protocol merges or reassigns the role under its rules and re-runs the full checks — solvability, containment, spoiler sweep, and tone sweep — plus the pack completeness check on any regenerated guide, before anything is re-issued. In Connected mode, invitations are sent only as the exact drafts you approved, one guest at a time; in Guide mode it hands you the finished text and you send it. You need: a designed, checked case (the other agents build it — or bring your own Case File) and your party's date, guest list, and casting notes.

How it works (about 5 minutes, once):

1. Add the agent file to your AI chat — upload the .md file, or on a phone, open the .txt version, copy it, and paste it in.

2. Type "Start".

3. Answer the agent's questions, one at a time. That's the calibration, and it's what makes the output fit YOUR party.

4. Done — ask for what you need.

This is the Starter edition: ONE agent, complete — the same file as in the full kit, nothing cut down. It includes the shared Case File and Party Profile (fill it once, paste it into any new chat, skip the interview), a prompt sheet with the exact phrases for this agent, a plain-language guide to what AI can and cannot do, and troubleshooting help.

Two ways to work — your choice, per session. Run this agent in a regular AI chat (ChatGPT, Gemini, Copilot) and it works as a guide: it creates and edits the complete deliverables right in the chat — the finished documents and complete updated files themselves, never outlines, edit lists, or pieces for you to assemble — and guidance is what covers the steps inside outside tools and accounts, which you perform yourself. Run the same file on an agentic AI — Muse, Claude Code, Devin, and similar — and it can do more: if you want, it walks you through connecting your tools and carries out its steps itself, showing you each consequential action for your approval first, verifying as it goes. Same agent, same rules — what changes is what your platform lets it do. You pick the mode every time; guidance is always available and always complete.

What's inside:

- The one agent, in .md AND .txt format (identical content)

- Prompt Sheet for this agent — the exact phrases for every job

- My Case File and Party Profile — the full profile; the sections for the other agents are already in it, so upgrading later means no new setup

- AI Limitations Disclaimer (.md + .txt)

- Starter README — step-by-step setup

- Troubleshooting & FAQ

What you need:

- An AI chat account of your own (free versions work for the guided work)

- Agent-specific notes are in the per-agent block below

Want the rest of the set later? The other three agents complete the kit (Standalone, $19). Starter owners upgrade for $10 — the difference — and the Case File and Party Profile you fill in here already covers the other agents, so there is no new setup.

Honest limits:

Each agent is a text instruction file for your existing AI — it is not a trained AI model, not a separate app, and not a boxed game. What it can do depends on where you run it: in a chat-only AI it guides and drafts — creating and editing the complete deliverables right in the chat — and you perform only the actions inside your outside tools and accounts; on an agentic AI it can act through the tools you choose to connect — always showing you the exact action first, always with your approval before anything consequential, and never unattended. Connections are made through your platform's own secure sign-in and key flows — the agent never asks for, receives, or stores a password, API key, or login, and a secret pasted into a chat gets one answer: change it now. This kit's spine holds in both modes: one shared Case File that no output may contradict, a solution deducible from the released clues alone, and the host-blind seal — structural in Connected mode (sealed content written directly to your folder, never displayed), a stated discipline in Guide mode (a clearly fenced block with save-without-reading instructions). Mode honesty holds too: a plain chat delivers print-ready content, layout specs, and the images themselves wherever it can generate them — an image prompt only as the labelled fallback where it cannot — never a claimed finished illustrated PDF. All characters are fictional, and your recorded tone — Family or Classic — is respected in every output. (Agent-specific limits are in the per-agent block below.)

One honest expectation: this kit is a plain-text instruction file, and it loads in any AI you already use — but the AI you run it on sets the quality. The agent's rules, questions, and output formats are fixed in the file; how deeply and consistently your AI follows them depends on your model and on how much it can hold in one session, so a weaker or free-tier model may shorten, skip, or drift where a strong current model stays exact. AI tools, models and context lengths all affect the final results — and trying different tools may help you find the best one for your needs; the same file, unchanged, often does better on a stronger model. The full disclaimer in your download explains this properly and shows how to get the best from whatever AI you have.

Price: $9 — Starter: ONE agent, your choice.