Services Agreement · Effective 2026-04-23
Services Agreement
Payment Terms
Client shall pay within 15 days of invoice.
The late fee is L = B × 0.015.
The document format for high-stakes work
Atto replaces Word for contracts, term sheets, and technical documents, bringing collaborative editing, mathematical notation, and binding execution into a single structured format.
Services Agreement · Effective 2026-04-23
Client shall pay within 15 days of invoice.
The late fee is L = B × 0.015.
---
title: Services Agreement
document_type: services_agreement
authoring_mode: hybrid
effective_date: 2026-04-23
---
## Payment Terms
Client shall pay within **15 days** of invoice.
The late fee is $L = B \times 0.015$.
::signature-block
- party: Baldwinson Corp.
signer: Curtis Baldwinson
title: CEO
sign_order: 1
:: The gap
A signature line in Word is just styled text. A tracked change is a cosmetic marker. A due date is a string. There is no semantic model under the surface; every meaning lives in the author’s head and the reviewer’s intuition.
That works for letters. It breaks for contracts, term sheets, research reports, and anything where the document itself is the event. There is no canonical source of truth, no clean audit trail, no way to prove the file you received is the file that was executed.
LaTeX solved notation and nothing else. No collaboration, no review workflow, no verification, no path from draft to executed document. Teams that need mathematical precision and counterparty screening end up stitching together a whole toolchain by hand, and reconciling every gap along the way.
Format: three layers, one file
Most formats solve one problem well and ignore the rest. Markdown markup language A plain-text writing format that uses simple symbols (like **bold** and # Heading) to mark structure and emphasis while staying readable as raw text. Created in 2004; widely used for documentation, blogs, and notes. is readable but has no notation. LaTeX
| lā-tek |
typesetting system A typesetting system designed for mathematical and scientific documents. Authors write formulas as code (e.g. \frac{a}{b} for a fraction) and the system renders them as properly typeset notation. The standard for academic publishing since the 1980s. has notation but requires a specialist to write it. Neither one can execute. Atto combines all three in a single plain-text file: Markdown carries the prose, LaTeX carries the math, and Atto carries the structure that makes it executable. The result is a document that a human can read, a machine can verify, and a court can rely on.
01
# Services Agreement
Between **Atto Technologies Inc.** ("Vendor") and
**Example Corp.** ("Customer"),
effective **April 23, 2026**.
## Scope
- Design consultation
- Proof of concept delivery
- Two rounds of revision
Payment terms are *net thirty*. Between Atto Technologies Inc. ("Vendor") and Example Corp. ("Customer"), effective April 23, 2026.
Payment terms are net thirty.
Markdown carries prose. It’s portable, diffable, and human-readable in any text editor. But it has no concept of notation, and no concept of meaning: a heading is just a heading, a signature line is just a bold word.
02
The present value of future cash flows is:
\[
PV = \sum_{t=1}^{T} \frac{C_t}{(1 + r)^t}
\]
where \(C_t\) is the cash flow in period
\(t\) and \(r\)
is the discount rate. The present value of future cash flows is:
PV = T Σ t=1 Ct (1 + r)twhere Ct is the cash flow in period t and r is the discount rate.
LaTeX renders notation the way a typesetter would. But the ecosystem was built for solo academic authors. No review threads, no signatures, no handoff to collaborators who don’t also write LaTeX.
03
1. Drafting
term-sheet.docx
2. Versioning
.git/
3. Notation
valuation.tex
4. Credit check
report.pdf
5. Signing
executed.pdf
6. Sharing
executed_FINAL.pdf
For leases, financing applications, and regulated agreements, every team assembles the same sprawling combination of accounts, platforms, and exports. The products change by industry and firm; the friction doesn’t. Every step rekeys data the last step already had. Every handoff produces a new file in a new format. And the executed PDF at the end (the one that gets filed, stored, and treated as the record of truth) has no structural relationship to the draft that was reviewed, the screening that approved it, or the math that priced it. It’s a photograph of a process that no longer exists anywhere.
04
# Services Agreement
Between **Atto Technologies Inc.** ("Vendor") and
**Example Corp.**, effective
**April 23, 2026**.
## Compensation
Vendor will be paid in quarterly installments, with net present value:
$$
PV = \sum_{t=1}^{T} \frac{C_t}{(1 + r)^t}
$$
::signature-block
party: Atto Technologies Inc.
signer: Curtis Baldwinson
title: CEO
sign_order: 1
:: Between Atto Technologies Inc. ("Vendor") and Example Corp., effective April 23, 2026.
Vendor will be paid in quarterly installments, with net present value:
PV = T Σ t=1 Ct (1 + r)t
One source file: Markdown prose, LaTeX notation, and typed directives,
all parsed to a single deterministic tree. The protocol layer makes
that tree do work: hash-chained revision history, signatures locked
to the content hash, executable math frozen at signing, and directives
like :::credit-check that gate execution on real-world
preconditions. All of it. In one file.
Collaboration: always in sync
Every section of an Atto document syncs the moment you touch it. Two authors can edit simultaneously with no locking, no “someone else is editing this,” and no overwritten work. A co-signer can execute their signature while a clause is still being revised on the other side of the document.
Drop your connection mid-edit. Come back. Everything is exactly where you left it.
Your browser
│
▼
Atto sync layer
│
▼
Document state ──── Change history
│
▼
Execution layer Workflow: start to signature
Every stage is a typed transition, captured in the document and the patch log. There is no external state.
Comparison
Atto is the only tool that does all of this. Here’s how the six it displaces compare across ten capabilities.
| Capability | Atto | Word / Google Docs | LaTeX / Markdown | Notion | DocuSign | Ironclad |
|---|---|---|---|---|---|---|
| Human-readable source format | Yes: native, built into the core product. | No: not supported. | Yes: native, built into the core product. | No: not supported. | No: not supported. | No: not supported. |
| Human-writable without special tools | Yes: native, built into the core product. | Partial: appended externally, lacks structural integrity, or requires another tool. 1 | Yes: native, built into the core product. | Partial: appended externally, lacks structural integrity, or requires another tool. 2 | No: not supported. | No: not supported. |
| Dedicated authoring editor | Yes: native, built into the core product. | Yes: native, built into the core product. | No: not supported. | Yes: native, built into the core product. | No: not supported. | Yes: native, built into the core product. |
| Hash-chained version history | Yes: native, built into the core product. | Partial: appended externally, lacks structural integrity, or requires another tool. 3 | Partial: appended externally, lacks structural integrity, or requires another tool. 4 | No: not supported. | No: not supported. | Partial: appended externally, lacks structural integrity, or requires another tool. 5 |
| Cryptographic version integrity | Yes: native, built into the core product. | No: not supported. | No: not supported. | No: not supported. | Partial: appended externally, lacks structural integrity, or requires another tool. 6 | Partial: appended externally, lacks structural integrity, or requires another tool. 5 |
| Signature execution | Yes: native, built into the core product. | No: not supported. | No: not supported. | No: not supported. | Yes: native, built into the core product. | Yes: native, built into the core product. |
| Signatures bound to content hash | Yes: native, built into the core product. | No: not supported. | No: not supported. | No: not supported. | Partial: appended externally, lacks structural integrity, or requires another tool. 7 | Partial: appended externally, lacks structural integrity, or requires another tool. 7 |
| Embedded math computation | Yes: native, built into the core product. | No: not supported. | Yes: native, built into the core product. | No: not supported. | No: not supported. | No: not supported. |
| Executable math (frozen at signing) | Yes: native, built into the core product. | No: not supported. | No: not supported. | No: not supported. | No: not supported. | No: not supported. |
| Native :::credit-check directive | Yes: native, built into the core product. | No: not supported. | No: not supported. | No: not supported. | No: not supported. | No: not supported. |