Workshops: Secure Agentic Coding & Governance
Open-Source Licensing & AI-Generated Code: Compliance Workshop
Coding agents occasionally produce code that closely matches public repositories, and a diff does not tell you where a function came from. This half-day workshop shows engineering leads, licence compliance owners and in-house counsel how to keep AI-generated code compatible with your open-source obligations and customer contracts, using scans, tool settings, contribution rules and documentation. We explain requirements and common practice; the legal assessment stays with your legal team.
Who is this workshop for?
- Engineering managers and heads of development
- Open source program offices and licence compliance owners
- In-house counsel and contract managers working with software
- Software vendors and agencies delivering source code to clients
- Developers contributing to external open-source projects
What you will take away
- Explain how licence risk arises with AI-generated code and how it differs from manual copying.
- Configure tool-side safeguards such as filters for matches with public code, where your tools offer them.
- Add snippet and licence scanning to your pipeline and handle matches through a defined process.
- Check customer contracts, vendor terms and indemnity commitments for clauses that affect AI-assisted delivery.
- Draft team rules for AI use in proprietary code and in contributions to open-source projects.
Agenda
Module 1
Where licence risk comes from
- Permissive and copyleft licences in brief: MIT, Apache-2.0, GPL, AGPL
- Reproduced snippets, generated boilerplate and newly added dependencies
- The open question of copyright in purely machine-generated code
Module 2
Safeguards in tools and pipeline
- Public-code filters and code referencing, for example in GitHub Copilot
- Snippet matching, licence scanning in SCA tools, SPDX and SBOM data
- Exercise: handling a scanner match from detection to decision
Module 3
Contracts and vendor terms
- Customer contracts: originality warranties, AI clauses, source code handover
- Terms for Claude, OpenAI Codex, GitHub Copilot and others: output rights, indemnities and their conditions
- Questions to raise with procurement and legal
Module 4
Rules for your team
- Building blocks for proprietary code and open-source contributions
- Recording AI assistance in commits and pull requests
- Exercise: drafting your one-page licensing rule set
A risk that is easy to overstate and easy to ignore
Large verbatim copies of licensed code from a modern coding agent are rare but not impossible, especially for well-known algorithms or very common files. Meanwhile agents add dependencies, copy example code from documentation and generate configuration from templates, each with its own licence. In daily practice, unchecked dependencies are often a bigger issue than memorised snippets.
We size the risk for your business model: a SaaS company has different obligations from a manufacturer shipping firmware or an agency handing over source code. With the AGPL in particular, that difference matters.
A note on legal advice
This workshop is not legal advice. We explain licence mechanics, tool settings and common industry practice; the legal assessment of your contracts and individual cases stays with your legal or compliance team, whom we are glad to involve from the start.
neooutput AI Code Engineers lead the technical part. The half day works best when engineering and legal attend together, remotely or on-site, in English or German.
Frequently asked questions
Can a coding agent bring GPL code into our proprietary product?
It can happen, usually through matching snippets or dependencies the agent adds. Licence scanners in the pipeline and filters in the tools reduce the risk considerably; a defined review process handles the remaining matches.
Do we own the code our agents write?
The business terms of the major vendors generally leave rights in the output with the customer. Whether purely machine-generated code is protected by copyright, however, depends on the jurisdiction and the human contribution. We explain the current debate; your legal team assesses the consequences.
May our developers contribute AI-assisted code to open-source projects?
That depends on the project. Some projects ask for disclosure or restrict AI-generated contributions, and sign-off processes such as the Developer Certificate of Origin ask contributors to confirm they have the right to submit. We help you align your internal rules with this.
Workshop
Get in touch
Should engineering and legal agree on one set of licensing rules for coding agents?
Tell us roughly how many people will join and pick dates that work – we will get back to you within 24 hours with a proposal for content, format and schedule.
Prefer to talk first?
15 minutes, free and without obligation – with our AI Trainers and AI Code Engineers.
Book a call for free (opens in new tab) +49 761 519 751 50 (24/7)Request a training / workshop
Workshop: Open-Source Licensing & AI-Generated Code: Compliance Workshop · Half day (approx. 4 hours)
More workshops
- Workshops: Secure Agentic Coding & GovernanceWriting a Coding Agent Policy: Workshop for Engineering Teams
- Workshops: Secure Agentic Coding & GovernanceAudit Trail for AI-Generated Code: Traceability Workshop
- Workshops: Secure Agentic Coding & GovernancePrivacy by Design with Coding Agents: GDPR Workshop for Developers
- Workshops: Secure Agentic Coding & GovernanceEU Cyber Resilience Act for Developers: Agentic Coding Workshop
- Workshops: Secure Agentic Coding & GovernanceNIS2 Workshop for Dev Teams: Secure Development with AI Agents
- Deploy & App StoresShipping a Claude Code App to the Apple App Store
