Freelancer Studio AI Starter — Follow-Up Drafter for a Queue and Drafts Built Only From Recorded Facts

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

One agent from the Freelancer Studio kit — complete, nothing cut down. Not a blank template and not a prompt pack: this agent fills your Studio Book with your own facts, gates its deliverable by name, and keeps that record current when your facts change. Works with the AI you already have: upload the file, tell it "Start", answer a few questions, and get to work.

One specialist freelancer paperwork agent — filling a faithful Studio Book, not handing you a blank template — 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 Follow-Up Drafter is the Book's voice, on a short leash (3 calibration questions): what follow-up you want drafted — for whom, and resting on which recorded facts (which invoice, scope, session, or status in the Book); what tone you want recorded, in your own words — a preference governs tone only and never supplies a fact, a promise, or a date; and, for each draft, what status you are reporting (prepared, approved, sent by you, dropped — in your words) and on what date you report it, as the standing question for every later status report.

Its follow-up queue runs a named completeness check (GATE-F) before it is presented as finished: one row per open item in the Book, none dropped, each carrying its recorded facts and its status as reported with its report date, ordered by recorded dates only — no urgency, likelihood, or "hot prospect" ordering exists in this kit — every prepared draft listed with the recorded facts it rests on, and the closing statement that no client is characterised in it and nothing in it was sent by the kit. A date passing changes no row: an invoice row does not become overdue where no due date is recorded, and a draft does not become sent.

In the kit's worked example, Sam's queue — a fictional example — holds exactly three open rows (her invoice of Monday 2027-03-08 at $1,116.25, payment not reported; her scope's $900 deposit per her own template, 50% × $1,800, received not reported; and the prospect's request for 20% off the logo package, recorded Tuesday 2027-02-23 as asked, with Sam's decision not recorded), and her draft of Wednesday 2027-03-10, in her recorded tone "short and friendly", rests only on the invoice facts — the invoice date, the three session dates and hours (11.75 hours at $95 an hour, totalling $1,116.25) — greeting no one by name because no contact person's name is recorded, and containing no due date, no delivery promise, and no discount, because none of those is recorded.

A request to just write something warm to chase a client with no recorded fact behind it is refused, with the intake route offered instead. Where a Gmail draft route verifies in a session, this agent may create its own draft in your own Gmail drafts after you approve that exact text — a draft, never a send: it sends nothing to anyone, in either mode, and it creates no calendar event and no Stripe draft (a recorded date's calendar copy is the Work Log & Invoice Keeper's work; this agent hands you the exact event block and names that route). If the Book's cover fields are missing, this agent names them and sends you to the Client Intake Coordinator first — the Client Intake Coordinator opens the Book every other agent reads. You need: the Book opened first (the Client Intake Coordinator, full kit / Upgrade, opens one — or bring a Studio Book you already have), the recorded facts each follow-up rests on, and your tone preference in your own words.

This kit builds and keeps a Studio Book from your own facts — it sends nothing, finalises nothing, collects nothing, negotiates nothing, scores nothing, drafts no legal language, and computes no tax. That boundary is part of the design of every agent in this kit, in this Starter as much as in the full kit: facts, figures, statuses, and notes are recorded only as you supply them, and a rate, term, figure, or date you have not supplied is printed as "To confirm" or "not recorded" — never invented, never estimated, never benchmarked.

How it works (one question at a time, 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 practice.
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 Studio Book (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.

Your Studio Book holds your prospects' and clients' names and the answers you report about them — other people's information — plus your rate card, your per-client pricing decisions, and unissued invoice and payment states. The agents never publish, post, or share your Book, and passwords, account numbers, payment details, and identity-document numbers are never recorded in the Book at all. Keep the Book where only the people you choose can see it, and share only the part a person needs.

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 where the platform genuinely allows: the Book kept as a file under git; and, for the agents that own them, calendar entries on your own calendar only, with no invitees by default, as the exact block you approved, and Stripe draft invoices in your own Stripe only where creating an invoice draft itself verifies as possible in that session — both created by the Work Log & Invoice Keeper only. Where a Gmail draft route verifies in a session, the Client Intake Coordinator or the Follow-Up Drafter may create its own draft in your own Gmail drafts after you approve that exact text — a draft, never a send. The Scope Builder touches no external route at all. 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. Every message is draft-only in both modes, no send route is claimed for any platform, and no route into FreshBooks, Wave, Bonsai, HoneyBook, or Dubsado exists in any mode: artefacts prepared for those systems are copy-ready, and carried in by you, stated plainly.

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 Studio Book — the full Book; 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, $49). Starter owners upgrade for $30 — the difference — and the Studio Book 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, not a bookkeeping service, and not a human freelancer service. Prompt packs for this work exist cheap, and generic freelancer prompts are easy to find free; blank freelancer templates are easy to find too. What this adds is the Book those products never build — filled only from your facts, its prices traced to your rate card, its figures recomputed with parts shown, its sessions reconciled, its deliverable gated, and the record kept current when your facts change — not a blank template or a bundle of prompts.

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 Studio Book that no output may contradict, facts recorded only as you supply them, unknowns printed as "To confirm", figures only as supplied and recomputed, no market rate or negotiated figure ever stated, no prospect scored, no agreement drafted or interpreted, no tax computed and no jurisdiction rule stated, statuses changed only on your report, and each main deliverable passing its named completeness gate against the recorded Book — a "just the highlights" version is labelled as an extract that failed the gate, never as the deliverable. (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: $19 — Starter: ONE agent, your choice.