docs: require agent, model, and settings in AI disclosure - #4512
Open
mnriem wants to merge 2 commits into
Open
Conversation
Contributor
There was a problem hiding this comment.
Copilot review overview
🟡 Changes recommended
Unresolved disclosure-policy consistency issues remain in CONTRIBUTING.md.
Once you've addressed the issues Copilot identified, you can request another Copilot review.
Review tier: Lite
Findings: 2
New issues introduced by this change (2)
| Severity | Finding |
|---|---|
 |
CONTRIBUTING.md — This policy says AI use must be disclosed in any issue, but the catalog-submission forms… |
|
CONTRIBUTING.md — The disclosure requirement lists four fields, including the extent of AI use, but this clause only… |
What changed in this PR
Updates AI-assistance disclosure requirements and adds structured disclosure fields to PR and issue templates.
Changes:
- Requires agent/tool, model(s), settings/mode, and extent disclosures.
- Adds structured fields to the pull request template.
- Adds required AI Disclosure fields to bug, feature, and agent issue forms.
| File | Review result |
|---|---|
CONTRIBUTING.md |
Requires changes: two moderate findings and one nit regarding disclosure consistency and guidance alignment. |
.github/PULL_REQUEST_TEMPLATE.md |
Reviewed; adds structured AI disclosure fields. |
.github/ISSUE_TEMPLATE/feature_request.yml |
Reviewed; adds a required AI disclosure field. |
.github/ISSUE_TEMPLATE/bug_report.yml |
Reviewed; adds a required AI disclosure field. |
.github/ISSUE_TEMPLATE/agent_request.yml |
Reviewed; adds a required AI disclosure field. |
Suppressed comments (1)
CONTRIBUTING.md:247
- Please update
AGENTS.md:525with this policy change. That repository guidance still says the model is optional (model, if applicable/<name-if-known>), so an agent following it can omit a model from a generated PR comment while violating the new hard requirement in this section. Keeping both versions makes the disclosure contract ambiguous.
If your PR responses or comments are being generated by an AI, disclose that as well, naming the same agent, model(s), and settings/mode.
💡 Configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
Make AI-assistance disclosure a hard requirement: when AI was used, the author must name the agent/tool, the model(s), the settings/mode (reasoning effort; autonomous vs. human-supervised), and the extent. This is stricter than the prior "model optional" stance and is a deliberate maintainer policy choice. It is framed as transparency, not a quality lever: disclosure does not change the level of scrutiny a contribution receives — the same evidence bar applies to everyone. It gives maintainers visibility into which models/settings are used and to what extent, which helps judge whether AGENTS.md guidance needs refinement over time. - CONTRIBUTING.md: upgrade the requirement, extend the PR responses/comments clause, refresh the example disclosures, and add the transparency rationale. Keep the trivial-fix exception. - PULL_REQUEST_TEMPLATE.md: add structured Agent/tool, Model(s), Settings/mode, and Extent fields prefilled with N/A. - bug_report, feature_request, agent_request issue forms: add a required AI Disclosure field defaulted to N/A. Assisted-by: GitHub Copilot (model: Claude Opus 4.8, autonomous) Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
mnriem
force-pushed
the
mnriem-docs-ai-disclosure-agent-model
branch
from
September 10, 2026 14:49
61cecf8 to
82e1fda
Compare
Contributor
There was a problem hiding this comment.
Copilot review overview
🟡 Changes recommended
Required disclosure dimensions remain incomplete in the guidance and PR template.
Once you've addressed the issues Copilot identified, you can request another Copilot review.
Review tier: Lite
Findings: 2
New issues introduced by this change (1)
| Severity | Finding |
|---|---|
 |
.github/PULL_REQUEST_TEMPLATE.md — The PR description calls for four structured disclosure fields (Agent/tool, Model(s),… |
Pre-existing issues (2)
| Severity | Finding |
|---|---|
|
CONTRIBUTING.md — The disclosure requirement lists four fields, including the extent of AI use, but this clause only… View comment |
|
CONTRIBUTING.md — This policy says AI use must be disclosed in any issue, but the catalog-submission forms… View comment |
Suppressed comments (2)
CONTRIBUTING.md:247
- The new hard-requirement wording for AI-generated PR responses/comments omits the extent of AI use, even though extent is one of the four required disclosure dimensions above and in the PR description. This leaves responses/comments documenting the agent, model, and mode but not whether AI drafted prose, code, or both.
If your PR responses or comments are being generated by an AI, disclose that as well, naming the same agent, model(s), and settings/mode.
CONTRIBUTING.md:258
- The detailed example omits an explicit autonomous-versus-human-supervised mode, even though that is now a required disclosure field. "Default settings" covers settings but not supervision, so copying this example still produces an incomplete disclosure.
> I consulted ChatGPT (GPT-5, default settings) to understand the codebase, but the solution
> was fully authored manually by myself.
Address review feedback on the AI-disclosure policy for consistency: - CONTRIBUTING.md: require the extent dimension in the AI-generated PR responses/comments clause too, so all four disclosure dimensions are listed consistently. - CONTRIBUTING.md: explicitly exempt the catalog-submission issue forms (extension/preset/bundle), which feed content-neutral validation automation and carry no disclosure field, so the "any issue" wording is no longer contradicted. - CONTRIBUTING.md: make the detailed example show the supervision mode. - AGENTS.md: drop "if applicable" from the PR-comment model disclosure so it matches the new hard requirement to name the model. Assisted-by: GitHub Copilot (model: Claude Opus 4.8, autonomous) Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Collaborator
Author
|
Addressed this review round in 7f2c8e1.
Disclosure: this comment and the commits in this round were produced by GitHub Copilot (model: Claude Opus 4.8), running autonomously under maintainer review, on behalf of @mnriem. |
Contributor
There was a problem hiding this comment.
Copilot review overview
🟡 Changes recommended
The updated AGENTS.md comment-disclosure example still uses “” and the sentence is awkwardly phrased, which weakens the intended “model required” policy language.
Once you've addressed the issues Copilot identified, you can request another Copilot review.
Review tier: Lite
Findings: 1
New issues introduced by this change (1)
| Severity | Finding |
|---|---|
|
AGENTS.md — This sentence reads awkwardly ("agent and model and the human") and the example still uses… |
Issues resolved since last review (3)
| Severity | Finding |
|---|---|
|
.github/PULL_REQUEST_TEMPLATE.md — The PR description calls for four structured disclosure fields (Agent/tool, Model(s),… View resolved comment |
|
CONTRIBUTING.md — The disclosure requirement lists four fields, including the extent of AI use, but this clause only… View resolved comment |
|
CONTRIBUTING.md — This policy says AI use must be disclosed in any issue, but the catalog-submission forms… View resolved comment |
| ### Comments | ||
|
|
||
| - If you are an agent working on behalf of a human, **disclose your identity in your PR comment** — name the agent (and model, if applicable) and the human you are acting for (e.g., "Posted on behalf of @user by GitHub Copilot (model: <name-if-known>)"). | ||
| - If you are an agent working on behalf of a human, **disclose your identity in your PR comment** — name the agent and model and the human you are acting for (e.g., "Posted on behalf of @user by GitHub Copilot (model: <name-if-known>)"). |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Makes AI-assistance disclosure a hard requirement: when AI was used, the contributor must name the agent/tool, the model(s), the settings/mode (reasoning effort; autonomous vs. human-supervised), and the extent (docs vs. code generation) — in the PR or issue.
This is intentionally stricter than the prior "model optional" stance and is a deliberate maintainer policy decision.
Framing (important)
This disclosure is transparency, not a quality lever. It does not change the level of scrutiny a contribution receives — the same evidence bar applies to everyone, human and AI-assisted alike. A strong PR from a weak model still merges; a broken PR from a frontier model is still rejected. The point is visibility into which models/agents/settings are in use and to what extent, which helps maintainers judge whether
AGENTS.mdguidance needs refinement over time. It is never a merge-quality prior or a brand-based gate. The existing trivial spacing/typo-fix exception is preserved.Changes
CONTRIBUTING.md(## AI contributions in Spec Kit): upgrade the requirement to name agent/tool + model(s) + settings/mode + extent; extend the AI-generated PR responses/comments clause to require the same (extent included); refresh both example disclosures; add the transparency (not scrutiny) rationale linking toAGENTS.md; update the "What we're looking for" disclosure bullet; explicitly exempt the catalog-submission forms. Trivial-fix exception kept..github/PULL_REQUEST_TEMPLATE.md: keep the did / did-not checkboxes and add a single required free-formAI disclosure:field pre-filled withN/A, with an HTML comment prompting for the agent/tool, model(s), settings/mode, and extent. A free-form field (rather than separate labeled fields) is a deliberate choice to keep the template light while still capturing all four dimensions..github/ISSUE_TEMPLATE/{bug_report,feature_request,agent_request}.yml: add a required free-form AI Disclosure field defaulted toN/A(issues are increasingly agent-authored too). Catalog-submission templates (extension,preset,bundle) are intentionally left untouched — they feed content-neutral validation automation and don't ask about AI use;CONTRIBUTING.mdnow states this exemption explicitly.AGENTS.md: drop "if applicable" from the PR-comment model-disclosure guidance so it matches the new hard requirement.Testing
markdownlint-cli2clean on the changed markdown (pre-existing fence findings excluded);git diff --checkclean; issue-form YAML parses and the new required field validates.Docs/policy-only change — no code paths touched.
AI Disclosure
AI disclosure: Drafted with GitHub Copilot (coding agent) using Claude Opus 4.8, autonomous mode under maintainer review; extent: prose and template edits from the maintainer's specification, with wording and framing reviewed by the maintainer before submission.