- Home
- Skills
- Business & Operations
- thats what she wrote
More screenshots
Works with the AI Agents you already use
thats what she wrote
Data-grounded client communication tool for writing accurate agency proposals, emails, and payment chasers.
$5
thats what she wrote
Example session with this skill installed
Write a quote for Acme Corp for a Website Audit based on our standard offers and their previous discount in projects.md.
- Read your context and instructions
- Compiled the thats what she
I have generated the quote for Acme Corp. I pulled the Website Audit rate from offers.md and applied the 10% loyalty discount found in their previous project record (ID: ACME-2023).
Proposal: Website Audit for Acme Corp
Client: Acme Corp
Date: October 24, 2023
Project Reference: ACME-2024-AUDIT
Scope of Work
Detailed technical review of the Acme Corp primary domain, including performance profiling, SEO crawl analysis, and security header audit as per our Standard Audit package.
Investment
| Item | Description | Price |
|---|---|---|
| Website Audit | Full technical site review | €1,200.00 |
| Loyalty Discount | 10% discount (applied from ACME-2023) | -€120.00 |
| Total | €1,080.00 |
Schedule & Next Steps
- Kickoff: (to confirm)
- Delivery: 10 business days from kickoff.
- Approval: Please sign and return this document to begin.
Check:
- €1,200.00: sourced from
offers.md(row: audit-std) - 10%: sourced from
projects.md(row: ACME-2023, field: discount) - €1,080.00: calculated (1200 * 0.9)
- Kickoff date: (to confirm) - No start date provided in request or
customers.md.
Next steps
- Provide the intended kickoff date to replace the placeholder.
- Run
node scripts/kit.mjs check draft.md ./datato verify the final file.
Connects securely to your tools. The creator never sees your data.
About this skill
Stop guessing prices and dates when writing for clients. This skill automates agency communications by grounding every word in your actual business data. It prevents the common AI habit of hallucinating discounts or project timelines by strictly referencing your local Markdown files.
What it does
- Generate proposals using structured data from your service catalog and project history.
- Draft email replies that answer client questions using existing customer records and project status.
- Automate chasers for late payments or missing approvals by checking current invoice data.
- Enforce house style by rewriting drafts to match your specific brand voice and formatting rules.
- Audit drafts for factual accuracy, flagging any figure or claim not found in your source files.
How it works
- Connect data by pointing the skill to your agency data folder containing files like
offers.mdandprojects.md. - Select task by asking to "quote for [client]", "answer this email", or "chase payment".
- Drafting occurs by merging your templates with real-time data while marking missing info as "(to confirm)".
- Verification runs a check against all prices and deliverables to ensure every amount has a valid source.
Frameworks & tools
The skill utilizes Markdown for data storage and includes a Node.js script (kit.mjs) for automated draft auditing and data validation.
Why this beats prompting it yourself
Generic prompts often invent professional-sounding but factually incorrect numbers. This skill forces the LLM to prove the source of every dollar amount and date, ensuring your proposals are always profitable and accurate.
Use cases
- Drafting a project estimate for a repeat client based on previous rates.
- Responding to a status query by pulling dates directly from your project tracker.
- Sending polite but firm payment reminders using your specific chaser templates.
- Auditing a manually written proposal to ensure it adheres to list prices.
Known limitations
Requires specific Markdown files to be present in the data folder. The amount checker script requires Node 18 or higher to run.
Requires structured Markdown data files. Audit script needs Node 18+. Cannot send emails directly. Missing data is marked with "(to confirm)" rather than estimated.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 20 days ago