What is GitHub Copilot CLI and how did the legal team use it?

GitHub Copilot CLI allows users to request tools and workflows in plain language without traditional coding. The legal team plugged it into their repositories and asked for what they wanted; Ngandu Kasuku used it to scaffold a contract drafting tool (terms-ai) with stored documents and instructions, while Jesse Geraci used it to create instruction sets and workflows for code analysis and DMCA triage that evolved into a desktop app.

How much time did these tools save?

Ngandu Kasuku cut his contract review and drafting time roughly in half since building terms-ai. Jesse Geraci's workflow tool expanded beyond its original DMCA focus to include contract review, NDA triage, risk assessment, and compliance checks, though no specific time savings are cited for his tool.

Are these tools open source?

terms-ai's tool and general workflow are open source, though the agreements and sensitive information are not part of the open source repository. Geraci's app requires some code and is not described as open source; it runs on a desktop interface with plain-language Markdown instructions.

AI Coding AssistantsAI Business & IndustryGitHub Blog (AI)Aug 4, 2026Aug 5, 2026, 06:00 JST

GitHub legal team builds AI tools without coding using Copilot CLI

Send on LINE

3 Key Points

What happened
GitHub's legal, program management, and business teams—not engineers—used GitHub Copilot CLI to build internal tools that automate repetitive work. Ngandu Kasuku, a product attorney, created terms-ai, a contract drafting tool that cut his review and drafting time roughly in half. Jesse Geraci, an online safety counsel, built a desktop app for analyzing source code and handling DMCA notices, expanding it to cover contract review, NDA triage, risk assessment, and compliance checks—all using plain-language instructions rather than traditional code.
Why it matters
The tools show that non-technical professionals can use AI to solve their own workflow problems without hiring engineers or waiting for vendor software. Both lawyers used GitHub Copilot CLI to turn repetitive, time-consuming tasks into structured, reusable processes—Kasuku by organizing documents and drafting resources in one repository, Geraci by encoding legal methodology into readable Markdown files that his team could edit and customize. This approach kept human judgment central while making workflows more consistent and scalable.
What to watch
The tools are designed for specific legal workflows but hint at a broader pattern: any job with repetitive, rule-based tasks (contract review, compliance checks, data analysis, triage) may be automatable without traditional software development. Both builders emphasized that these are decision-support systems, not replacements for professional judgment—Kasuku noted his tool still has room to improve, and Geraci stressed that his legal Copilot "shouldn't be treated" as a replacement for legal review.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, in your inbox every morning.

Get news like this every morning, free

Context & Analysis

The article presents a case study in how non-technical professionals are using AI tooling to solve their own workflow problems. Both Ngandu Kasuku and Jesse Geraci came to GitHub Copilot CLI with a narrow, repetitive pain point—contract drafting and DMCA analysis, respectively—and discovered that they could build lasting solutions without engineering support. The key difference between their initial approaches and their final tools was architectural: Kasuku moved from one-off prompts to a single repository that centralized documents, instructions, and resources, which reduced copying and pasting and made drafting more consistent. Geraci similarly transformed ad-hoc prompts into structured instruction sets (written in plain language and Markdown) that encoded his legal methodology—including different analysis modes for clients versus lawyers, external data integration, and escalation routing—all without traditional code. Both stressed that their tools are decision-support systems, not replacements for human judgment; Kasuku acknowledged that terms-ai "still has a long way to go," and Geraci explicitly cautioned that his tool "shouldn't be treated" as a replacement for legal review. This positioning matters because it suggests that the value of these tools lies not in automation per se, but in making professional workflows more consistent, transparent, and scalable while keeping human expertise central.

FAQ

What is GitHub Copilot CLI and how did the legal team use it?: GitHub Copilot CLI allows users to request tools and workflows in plain language without traditional coding. The legal team plugged it into their repositories and asked for what they wanted; Ngandu Kasuku used it to scaffold a contract drafting tool (terms-ai) with stored documents and instructions, while Jesse Geraci used it to create instruction sets and workflows for code analysis and DMCA triage that evolved into a desktop app.
How much time did these tools save?: Ngandu Kasuku cut his contract review and drafting time roughly in half since building terms-ai. Jesse Geraci's workflow tool expanded beyond its original DMCA focus to include contract review, NDA triage, risk assessment, and compliance checks, though no specific time savings are cited for his tool.
Are these tools open source?: terms-ai's tool and general workflow are open source, though the agreements and sensitive information are not part of the open source repository. Geraci's app requires some code and is not described as open source; it runs on a desktop interface with plain-language Markdown instructions.

GitHub Blog (AI)Read Original Article

Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Get it free

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.