Repository navigation
Add AI usage guideline - #851
Conversation
Define how the Eclipse Generative AI Usage Guidelines, made applicable by wp__policies, are applied in S-CORE: contributor responsibility, mandatory disclosure (Assisted-by trailer in commits and AI disclosure section in the PR description), file header for largely AI-generated files, human approval for every merge and the relation to functional safety and tool management. Relates to eclipse-score#189 Assisted-by: Anthropic Claude Opus 5.5 (claude-opus-5-5) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
The created documentation from the pull request is available at: docu-html |
|
|
||
| This guideline defines how the | ||
| `Eclipse Generative AI Usage Guidelines <https://www.eclipse.org/projects/guidelines/genai/>`_, | ||
| made applicable by :need:`wp__policies`, are applied in *S-CORE*. It applies to every |
There was a problem hiding this comment.
I think we do not use S-CORE in the process description, project independent, e.g. compare your update in the work product below, "Their application in the project"
There was a problem hiding this comment.
Agreed, changed to "the project" throughout the guideline.
Assisted-by: Anthropic Claude Opus 5.5 (claude-opus-5-5)
| This guideline defines how the | ||
| `Eclipse Generative AI Usage Guidelines <https://www.eclipse.org/projects/guidelines/genai/>`_, | ||
| made applicable by :need:`wp__policies`, are applied in *S-CORE*. It applies to every | ||
| contribution to any *S-CORE* repository (source code, documentation, process description, |
There was a problem hiding this comment.
the list seems not complete, may better to link to the repo overview of https://github.com/orgs/eclipse-score/repositories
There was a problem hiding this comment.
Removed the enumeration instead of linking a repository list, so the guideline stays project independent: it now applies to every contribution to the project, independent of the repository and the kind of content, including issues, pull requests and review comments.
Assisted-by: Anthropic Claude Opus 5.5 (claude-opus-5-5)
| defend it in the review. | ||
| #. The use of a generative AI tool shall not reduce the scope of the review, the acceptance | ||
| criteria or the automated checks that apply to a contribution. | ||
| #. The contributor shall follow the terms of use of the AI tool and the policy of their |
There was a problem hiding this comment.
Not every contributer may an employee, so is this not something we should left out. If someone has to follow, would be anyhow covered from company policies
There was a problem hiding this comment.
Agreed, removed. It is covered by the Eclipse guidelines anyway.
Assisted-by: Anthropic Claude Opus 5.5 (claude-opus-5-5)
| criteria or the automated checks that apply to a contribution. | ||
| #. The contributor shall follow the terms of use of the AI tool and the policy of their | ||
| employer on the use of generative AI. | ||
| #. The contributor shall not enter confidential information, personally identifiable |
There was a problem hiding this comment.
Would also not put that here, we are open source, it is the responsibility and such rules should on ECLIPSE, if required
There was a problem hiding this comment.
Agreed, removed. Same reason, the Eclipse guidelines already cover it.
Assisted-by: Anthropic Claude Opus 5.5 (claude-opus-5-5)
|
|
||
| #. The disclosure shall be given in both of the following places: | ||
|
|
||
| - in every affected commit, as ``Assisted-by:`` trailer in the commit message, and |
There was a problem hiding this comment.
are two places not to much, can we brake it down to one
There was a problem hiding this comment.
Agreed, reduced to one place: the disclosure is given where the AI-assisted content is created. For changes in a repository this is the Assisted-by: trailer in every affected commit (squash merge keeps the commit messages in this repo). For issues, PR descriptions and comments it is an Assisted-by: line in that text.
Assisted-by: Anthropic Claude Opus 5.5 (claude-opus-5-5)
| If several AI tools were used, one ``Assisted-by:`` line per tool shall be given. | ||
|
|
||
| #. A file that is largely or entirely AI-generated shall carry, in addition to the standard | ||
| *S-CORE* copyright header, an AI disclosure block directly below it: |
There was a problem hiding this comment.
Changed to "copyright header of the project".
Assisted-by: Anthropic Claude Opus 5.5 (claude-opus-5-5)
| branches or merge pull requests. | ||
| #. A committer may reject an AI-assisted contribution that they do not understand or are not | ||
| willing to maintain, or request that it is split into smaller pull requests. | ||
| #. A committer may request the missing disclosure, if a contribution appears to be AI-assisted |
There was a problem hiding this comment.
Changed to "shall request", consistent with the mandatory disclosure.
Assisted-by: Anthropic Claude Opus 5.5 (claude-opus-5-5)
| process, review and verification requirements as any other content. No additional rules | ||
| apply. | ||
| #. The output of an AI tool shall not be used as verification evidence on its own. If AI | ||
| output becomes part of a work product without a complete human review, the AI tool shall be |
There was a problem hiding this comment.
What is meant here, automated verification evidence without human review, shall not allowed, that it is. The link to Tool Evaluation is generic, AI will not be expected to be qualified according functional safety
There was a problem hiding this comment.
Reworded: AI tools used to create or verify work products are identified and evaluated like any other tool according to Tool Management. A qualification of AI tools is not intended, therefore AI output shall not be used as work product or verification evidence without human review.
Assisted-by: Anthropic Claude Opus 5.5 (claude-opus-5-5)
There was a problem hiding this comment.
🟡 Changes recommended
The guideline’s disclosure rule cannot accommodate standalone issues and comments, and its Tool Management condition could incorrectly exempt fully reviewed AI output from required identification and confidence assessment.
2 open issues
What changed in this PR
Defines how S-CORE applies the Eclipse Generative AI Usage Guidelines, addressing issue #189 and integrating the guidance into the existing process-management documentation.
Changes:
- Documents contributor responsibility, AI disclosure requirements, and headers for largely AI-generated files.
- Defines human merge approval requirements and the relationship to safety, security, and Tool Management.
- Links the guideline from project policies, the process-management workflow, and the guidance index.
| File | Description |
|---|---|
process/process_areas/process_management/process_management_workproducts.rst |
Links project policies to the AI usage guideline. |
process/process_areas/process_management/process_management_workflow.rst |
Includes the guideline in the process-management workflow. |
process/process_areas/process_management/guidance/index.rst |
Adds the guideline to documentation navigation. |
process/process_areas/process_management/guidance/ai_usage_guideline.rst |
Introduces project-wide AI usage and disclosure requirements. |
Review effort: Balanced
| corrections, formatting, code completion of single statements or rote boilerplate that | ||
| the contributor modifies and supplements. | ||
|
|
||
| #. The disclosure shall be given in both of the following places: |
There was a problem hiding this comment.
Addressed together with the comment from masc2023: disclosure is now given where the content is created, i.e. commit trailer for repository changes and an Assisted-by: line directly in issues, PR descriptions and comments.
Assisted-by: Anthropic Claude Opus 5.5 (claude-opus-5-5)
| #. The output of an AI tool shall not be used as verification evidence on its own. If AI | ||
| output becomes part of a work product without a complete human review, the AI tool shall be | ||
| handled by Tool Management (:need:`wf__tool_evaluate_tool`). |
There was a problem hiding this comment.
Addressed: AI tools are now identified and evaluated according to Tool Management regardless of the human review. The human review is required in addition, because a qualification of AI tools is not intended.
Assisted-by: Anthropic Claude Opus 5.5 (claude-opus-5-5)
- Use project independent wording instead of S-CORE - Remove rules already covered by the Eclipse guidelines (employer policy, confidential data) - Disclose AI use once, where the content is created (commit trailer or Assisted-by line in issue, PR description or comment) - Committer shall request missing disclosure - Clarify relation to Tool Management: AI tools are identified and evaluated, qualification is not intended, human review is required Assisted-by: Anthropic Claude Opus 5.5 (claude-opus-5-5) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Define how the Eclipse Generative AI Usage Guidelines, made applicable by wp__policies, are applied in S-CORE: contributor responsibility, mandatory disclosure (Assisted-by trailer in commits and AI disclosure section in the PR description), file header for largely AI-generated files, human approval for every merge and the relation to functional safety and tool management.
Relates to #189
Assisted-by: Anthropic Claude Opus 5.5 (claude-opus-5-5)