Add ReEDS AI Policy by kennedy-mindermann · Pull Request #237 · ReEDS-Model/ReEDS

kennedy-mindermann · 2026-10-01T15:49:52Z

Summary

This PR adds an AI policy for ReEDS to the contributing guide. It also adds an AGENTS.md file into the repo.

Technical details

Implementation notes

The AGENTS.md file is from Wesley.

Additional changes

Switches added/removed/changed

Issues resolved

Known incompatibilities

Relevant sources or documentation

Validation, testing, and comparison report(s)

Checklist for author

Details to double-check

General information to guide review

Zero impact on results of default case
No large data file(s) added/modified
No substantive impact on runtime for full-US reference case
No substantive impact on folder size for full-US reference case
No change to process flow (runreeds.py, reeds/core/solve/solve.py)
No change to code organization
No change to package requirements (environment.yml or Project.toml)

Did you use LLM tools (chatbot or copilot) in the preparation of this PR? If so, describe how

No, I didn't use AI when writing the AI policy.

Tag points of contact here if you would like additional review of the relevant parts of the model

pesap · 2026-10-06T20:01:54Z

@kennedy-mindermann , I like were this is going. Do you think it would be worth it to add to the AGENTS.md that it needs to match the code style of the repo? I have found that if you do not specify it, it will mix and match any style that it finds convenient at a time.

patrickbrown4 · 2026-10-08T22:41:47Z

@@ -0,0 +1,467 @@
+# AGENTS.md — Guidance for AI Coding Assistants
+
+This file gives AI coding tools (Claude, ChatGPT, GitHub Copilot, Cursor, Codex,


(nitpick) I think we can avoid length-based line breaks in markdown files; since single breaks don't have semantic value, you can view these files with wrapping, and length-based breaks make later edits annoying (if you change the first line of a paragraph, you don't want to redo the spacing for every line that follows). In the docs we break on sentence-ending periods (and other punctuation for long sentences), so I'd do the same here.

patrickbrown4 · 2026-10-08T22:42:42Z

@@ -0,0 +1,467 @@
+# AGENTS.md — Guidance for AI Coding Assistants
+
+This file gives AI coding tools (Claude, ChatGPT, GitHub Copilot, Cursor, Codex,


Do we need to list commercial products by name? If not, I'd avoid it to avoid an appearance of endorsement.

patrickbrown4 · 2026-10-08T22:43:56Z

+# AGENTS.md — Guidance for AI Coding Assistants
+
+This file gives AI coding tools (Claude, ChatGPT, GitHub Copilot, Cursor, Codex,
+etc.) the context and operating rules they need to make safe, high-quality


I don't think this file ensures the contributions are safe or high-quality. Could probably just cut this sentence.

patrickbrown4 · 2026-10-08T22:46:34Z

+  requires a commercial license (CPLEX or compatible); Python 3.11 drives
+  preprocessing, orchestration, and postprocessing; Julia 1.12.1


We use python 3.14 now (see environment.yaml). Do we have to specify the version here?

Nop. Actually, we should point it to the environment as you mentioned. Cheaper models will not pick-it up right away.

patrickbrown4 · 2026-10-08T22:48:30Z

@Max-Vanatta you also had a version of this file, right?

Yep, I have one that does much of the same but for general coding rather than ReEDS specific. Happy to draft a version of this file with any of those additions are valuable to the team.

patrickbrown4 · 2026-10-09T15:36:05Z

+the report, etc. You can re-run any individual step in isolation against the
+inputs/outputs the prior steps already wrote into the run folder. This is
+much faster than a full rerun and is often the right first check while
+iterating on a single file.


It depends, there are still a few cases where files are overwritten in place (e.g. in forecast.py)

patrickbrown4 · 2026-10-09T15:39:31Z

+  [`reeds/input_processing/runfiles.csv`](reeds/input_processing/runfiles.csv)
+  to identify which files a given script writes.


runfiles.csv mostly only has the directly copied files; it doesn't include most files written by input_processing scripts

patrickbrown4 · 2026-10-09T15:44:25Z

+* **DON'T** use AI to write PR descriptions.
+* **DON'T** use AI-generated text in `model_documentation.md`. All text there must be original. *See below for permitted AI uses in this file*.
+* **DO** disclose AI use in the PR template's LLM section, including *how* you verified any LLM-generated content.
+* **DO** include verification for PRs with meaningful AI contributions so the reviewer can confirm correctness without rederiving the change themselves.


Could we be more specific on what "verification" means here?
For example, I'd support always running USA_defaults for LLM-heavy PRs.

patrickbrown4 · 2026-10-09T15:47:45Z

+### Recommended Practices
+**Planning and scope**
+* Start in plan mode, and include a verification step in the plan.
+* Break larger efforts into multiple PRs.


PR should still stand alone though; it'd be good to avoid churn and temporary code on the main branch.
Consider using the approach for #185, where the subsidiary PRs are being merged into wjc/dev/separate_class instead of main, and that branch will then be compared and reviewed again before merging.

patrickbrown4 · 2026-10-09T15:49:21Z

+* Have the AI tool pause at checkpoints for human review before proceeding.
+
+**While developing**
+* If you don't understand why a change was made, dig into it before moving on.


It says above that "Contributors are responsible for understanding and verifying every change"; I'd use similar language here

add draft of AI policy

a6ef8f3

github-actions Bot added the docs label Oct 1, 2026

Merge branch 'main' into km/ai-policy

3d68194

kennedy-mindermann added 2 commits October 6, 2026 16:26

add exception to agents.md file to not edit model_documentation.md

8a20f43

Merge branch 'main' into km/ai-policy

0142f8d

patrickbrown4 self-requested a review October 7, 2026 19:10

Merge branch 'main' into km/ai-policy

2bb794d

patrickbrown4 reviewed Oct 9, 2026

View reviewed changes

kennedy-mindermann added 3 commits October 20, 2026 09:17

modify agents file to accomodate PR review comments

177a757

fix spacing

7bdcb0d

address more PR comments

77181c2

kennedy-mindermann requested a review from patrickbrown4 October 20, 2026 15:17

address PR comment

42dda94

		@@ -0,0 +1,467 @@
		# AGENTS.md — Guidance for AI Coding Assistants

		This file gives AI coding tools (Claude, ChatGPT, GitHub Copilot, Cursor, Codex,

		requires a commercial license (CPLEX or compatible); Python 3.11 drives
		preprocessing, orchestration, and postprocessing; Julia 1.12.1

		[`reeds/input_processing/runfiles.csv`](reeds/input_processing/runfiles.csv)
		to identify which files a given script writes.

Conversation

kennedy-mindermann commented Oct 1, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Technical details

Implementation notes

Additional changes

Switches added/removed/changed

Issues resolved

Known incompatibilities

Relevant sources or documentation

Validation, testing, and comparison report(s)

Checklist for author

Details to double-check

General information to guide review

Did you use LLM tools (chatbot or copilot) in the preparation of this PR? If so, describe how

Tag points of contact here if you would like additional review of the relevant parts of the model

Uh oh!

pesap commented Oct 6, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

This branch has not been deployed

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

kennedy-mindermann commented Oct 1, 2026 •

edited

Loading