Skip to content

docs(ai-policy): allow AI-drafted text written for human readers - #839

Merged
nickvergessen merged 1 commit into
masterfrom
docs/ai-policy-human-readable-text
Nov 17, 2026
Merged

nickvergessen merged 1 commit into
masterfrom
docs/ai-policy-human-readable-text

Conversation

@AndyScherzinger

@AndyScherzinger AndyScherzinger commented Nov 16, 2026 •

Copy link
Copy Markdown
Member

The AI policy requires issues, PR descriptions and review comments to be in the contributor's own words, so agent-drafted text has to be retyped by hand. Following nextcloud/mail#14132, this allows AI-drafted text written for human readers and judges contributions by their result, not by how they were produced.

  • AI-drafted text must be plain, disclosed, and follow the repo's templates. PR descriptions say what problem they solve, what changed, and how to verify it; bug reports state whether the bug was reproduced on a live instance.
  • Maintainers may ask for a rewrite before closing a PR that shows neither understanding nor quality. PRs opened before a policy change are asked to adapt.
  • Every PR that changes runtime behavior is tested on a live instance, with AI help if wanted, but not only by AI.
  • Commit message bodies explain why, not how.
  • The snippet's "never do" list now reads "Don't X. Instead, Y." so smaller models know what to do, and the Assisted-by examples name models that exist.

Co-authored by @MiMoHo, whose #818 is the source of the rewrite and grandfathering rules.

🤖 AI (if applicable)

  • The content of this PR was partly or fully generated using AI

@nextcloud-command nextcloud-command added the AI assisted This PR contains AI-assisted commits label Nov 16, 2026
@AndyScherzinger
AndyScherzinger force-pushed the docs/ai-policy-human-readable-text branch 5 times, most recently from 7a0b739 to 503598e Compare November 17, 2026 12:37
@AndyScherzinger
AndyScherzinger marked this pull request as ready for review November 17, 2026 12:41
@AndyScherzinger
AndyScherzinger force-pushed the docs/ai-policy-human-readable-text branch from 503598e to fb7471d Compare November 17, 2026 15:53
@AndyScherzinger
AndyScherzinger force-pushed the docs/ai-policy-human-readable-text branch 4 times, most recently from 51430a1 to e4142f8 Compare November 17, 2026 18:38
The policy required issues, PR descriptions and review comments to be
in the contributor's own words, so agent-drafted text had to be
retyped by hand. Allow AI-drafted text written for humans and judge the
result: maintainers ask for a rewrite before closing, and PRs opened
before a policy change are asked to adapt rather than closed.

Ask for PR descriptions that answer the reviewer's first questions,
reproducible bug reports, commit messages that explain why, and live
testing for every runtime change. Rewrite the snippet's "never do" list
as "don't X, instead Y" so smaller models can follow it, and replace
example models that are retired or never existed.

Co-authored-by: MiMoHo <37556964+MiMoHo@users.noreply.github.com>
Assisted-by: ClaudeCode:claude-opus-5-5
Signed-off-by: Andy Scherzinger <info@andy-scherzinger.de>
@AndyScherzinger
AndyScherzinger force-pushed the docs/ai-policy-human-readable-text branch from e4142f8 to b4317a2 Compare November 17, 2026 18:38
@nickvergessen
nickvergessen merged commit 71fa031 into master Nov 17, 2026
6 checks passed
@nickvergessen
nickvergessen deleted the docs/ai-policy-human-readable-text branch November 17, 2026 20:27
ChristophWurst added a commit to nextcloud/mail that referenced this pull request Nov 18, 2026
Follows nextcloud/.github#839. The old "own words" rule forced
agent-drafted PR and issue text to be retyped by hand. The policy now
judges contributions by their result and requires text that is plain,
disclosed and written for human readers.

Intentionally left out: the upstream AI_POLICY.md and snippet files are
not copied; only the parts that apply to this repo's own docs are
ported. Agents may sign off with `git commit -s`, but only after the
contributor has reviewed and approved the changes, which is a
deliberate deviation from the upstream snippet.

Assisted-by: ClaudeCode:claude-sonnet-5-5
Signed-off-by: Christoph Wurst <1374172+ChristophWurst@users.noreply.github.com>
GVodyanov added a commit to whoalin1/contacts that referenced this pull request Nov 18, 2026
The org AI policy (nextcloud/.github#839) now allows issues, PR
descriptions and comments to be drafted with AI, as long as they are
written for people, disclosed and follow the templates. It also asks
for live-instance testing of runtime changes and commit bodies that
explain why rather than how. Bring AGENTS.md in line so agents follow
the current policy.

Assisted-by: ClaudeCode:claude-opus-5-5
Signed-off-by: Grigory V <scratchx@gmx.com>
GVodyanov added a commit to abhinavohri/calendar that referenced this pull request Nov 18, 2026
The org AI policy (nextcloud/.github#839) now allows issues, PR
descriptions and comments to be drafted with AI, as long as they are
written for people, disclosed and follow the templates. It also asks
for live-instance testing of runtime changes and commit bodies that
explain why rather than how. Bring AGENTS.md in line so agents follow
the current policy.

Assisted-by: ClaudeCode:claude-opus-5-5
Signed-off-by: Grigory V <scratchx@gmx.com>
This was referenced Nov 21, 2026
enoch85 pushed a commit to nextcloud/vm that referenced this pull request Nov 21, 2026
…#2896)

The org-wide AI policy now allows AI-drafted issues, PR descriptions
and comments as long as they are written for human readers, disclosed
and posted on the contributor's behalf. The agent instructions still
told agents never to write such text, which contradicts the policy
they point to.

Runtime changes must also be validated on a live instance, and agent
testing may support but not replace the contributor's own testing.

Assisted-by: ClaudeCode:claude-opus-5-5

Signed-off-by: Andy Scherzinger <info@andy-scherzinger.de>
AndyScherzinger added a commit to nextcloud/server that referenced this pull request Nov 21, 2026
…b#839

The contributing guide still required PR descriptions and comments in
the contributor's own words and live testing only for new features.
Bring its AI section in line with the org-wide contributing guide.

Assisted-by: ClaudeCode:claude-opus-5-5
Signed-off-by: Andy Scherzinger <info@andy-scherzinger.de>
AndyScherzinger added a commit to nextcloud/desktop that referenced this pull request Nov 21, 2026
…b#839

The contributing guide still required PR descriptions and comments in
the contributor's own words and live testing only for new features.
Bring its AI section in line with the org-wide contributing guide.

Assisted-by: ClaudeCode:claude-opus-5-5
Signed-off-by: Andy Scherzinger <info@andy-scherzinger.de>
AndyScherzinger added a commit to nextcloud/all-in-one that referenced this pull request Nov 21, 2026
…b#839

The contributing guide still required PR descriptions and comments in
the contributor's own words and live testing only for new features.
Bring its AI section in line with the org-wide contributing guide.

Assisted-by: ClaudeCode:claude-opus-5-5
Signed-off-by: Andy Scherzinger <info@andy-scherzinger.de>
AndyScherzinger added a commit to nextcloud/context_chat that referenced this pull request Nov 21, 2026
…ub#839

The repository carries its own copy of the AI policy and of the
contributing guide, both still requiring PR descriptions and comments
in the contributor's own words and live testing only for new
features. Sync both with the current org-wide versions.

Assisted-by: ClaudeCode:claude-opus-5-5
Signed-off-by: Andy Scherzinger <info@andy-scherzinger.de>
AndyScherzinger added a commit to nextcloud/context_chat_backend that referenced this pull request Nov 21, 2026
…ub#839

The repository carries its own copy of the AI policy and of the
contributing guide, both still requiring PR descriptions and comments
in the contributor's own words and live testing only for new
features. Sync both with the current org-wide versions.

Assisted-by: ClaudeCode:claude-opus-5-5
Signed-off-by: Andy Scherzinger <info@andy-scherzinger.de>
AndyScherzinger added a commit to global-prog/forms that referenced this pull request Nov 21, 2026
The org-wide AI policy now allows AI-drafted issues, PR descriptions
and comments as long as they are written for human readers, disclosed
and posted on the contributor's behalf. Add the rules this file was
missing, such as commit bodies that explain why and live-instance
validation, and drop those that contradict the policy.

Assisted-by: ClaudeCode:claude-opus-5-5
Signed-off-by: Andy Scherzinger <info@andy-scherzinger.de>
UlyssesZh pushed a commit to UlyssesZh/spreed that referenced this pull request Nov 21, 2026
The org-wide AI policy now allows AI-drafted issues, PR descriptions
and comments as long as they are written for human readers, disclosed
and posted on the contributor's behalf. Add the rules this file was
missing, such as commit bodies that explain why and live-instance
validation, and drop those that contradict the policy.

Assisted-by: ClaudeCode:claude-opus-5-5
Signed-off-by: Andy Scherzinger <info@andy-scherzinger.de>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

AI assisted This PR contains AI-assisted commits

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants