@ggdna/dna-gg 0.5.1 → 0.7.0

package/README.md

@@ -32,11 +32,6 @@ All skills of this layer are prefixed with `gg-`:
 - `/gg-publish` — releases the ticket
 - `/gg-cleanup` — removes what the ticket left behind
 
-## Configuration
-
-- `dna/dot-github/workflows/quick_check.yaml` — the quick check pipeline
-  every pull request has to pass
-
 ## Scripts
 
 - `dna/scripts/delete-feature-branch.js` — the one script this layer calls

package/dna/_dna.json

@@ -5,12 +5,13 @@
 
   // The parent DNAs. dna_install brings the install overview this layer
   // extends with the gg install guides, dna_index the index conventions
-  // every gg repo follows. Parents first, this layer wins.
+  // every gg repo follows, dna_github the GitHub repo setup and the quick
+  // check workflow. Parents first, this layer wins.
   //
   // dna_scripts is deliberately not a parent: this layer calls exactly one
   // of its scripts — `delete-feature-branch.js` from the /gg-cleanup skill —
   // and carries its own copy of it under dna/scripts/ instead. A consumer
   // that wants the full script set lists dna_scripts itself; the copies
   // here are byte-identical, so having both layers changes nothing.
-  "layers": ["dna_install", "dna_index"]
+  "layers": ["dna_install", "dna_index", "dna_github"]
 }

package/dna/_generated.json

@@ -25,13 +25,21 @@
       "via": null,
       "hash": "0x1d92b0add37a1ac7"
     },
+    {
+      "name": "dna-github",
+      "package": "@ggdna/dna-github",
+      "ecosystem": "node",
+      "resolvedVersion": "0.0.2",
+      "via": null,
+      "hash": "0xae24c13573219f91"
+    },
     {
       "name": "self",
       "package": null,
       "ecosystem": null,
       "resolvedVersion": null,
       "via": null,
-      "hash": "0x80c6dc9dbb7e23c3"
+      "hash": "0xc0e2cebcfa39b9d7"
     }
   ],
   "claude": {
@@ -47,11 +55,11 @@
     },
     {
       "path": ".claude/skills/gg-commit/SKILL.md",
-      "hash": "0xd2e82c49f737093b"
+      "hash": "0xdb7a268397facfc4"
     },
     {
       "path": ".claude/skills/gg-publish/SKILL.md",
-      "hash": "0xaa7d82e9b407f561"
+      "hash": "0x61626e5e2425d9ab"
     },
     {
       "path": ".claude/skills/gg-push/SKILL.md",
@@ -59,7 +67,7 @@
     },
     {
       "path": ".claude/skills/gg-ticket/SKILL.md",
-      "hash": "0x06f83f4f73c9fb42"
+      "hash": "0xe2e320fd71159027"
     },
     {
       "path": ".claude/skills/gg/SKILL.md",
@@ -87,7 +95,7 @@
     },
     {
       "path": "doc/guides/for-ai/ai-dev-guide.md",
-      "hash": "0xbbaf82081767678b"
+      "hash": "0xa7d0d75d19ccfb0f"
     },
     {
       "path": "doc/guides/index-guide.md",
@@ -144,6 +152,14 @@
     {
       "path": "doc/templates/index-template.jsonc",
       "hash": "0x16c3a8d50c72ed6d"
+    },
+    {
+      "path": "guides/setup-github-guide.md",
+      "hash": "0x8681f11bbfaa0c00"
+    },
+    {
+      "path": "scripts/setup-github-repo.js",
+      "hash": "0xc05910cfe214fe3a"
     }
   ]
 }

package/dna/doc/guides/for-ai/ai-dev-guide.md

@@ -62,8 +62,16 @@ repositories. The ticket then carries no plan of its own.
 
 ## Add git repositories
 
-Look at the `index.jsonc` of each repo in .ocean and decide which repos need to
-be added to the ticket.
+Ask the user how the repos are chosen, as a single question with two
+options: automatically or manually.
+
+Manually: ask in a second single input field which repos the ticket needs,
+e.g. `repo1 repo2`. Check each name against the repos in .ocean and ask
+again for names that do not exist there. The entered list is the
+confirmation; add the repos as shown below.
+
+Automatically: look at the `index.jsonc` of each repo in .ocean and decide
+which repos need to be added to the ticket.
 Make a plan for how you roughly want to implement the ticket.
 If certain parts of the implementation belong to a domain that does not yet
 exist in the .ocean folder, consider creating a new repository.
@@ -106,6 +114,42 @@ ticket.
 - Plan only: skip this step. The following steps then apply to the project
   management repo only.
 
+## Keep the publish config up to date
+
+Each repo of the ticket has a file `.gg/publish_config.json`. gg offers its
+values as defaults: `gg do commit` pre-fills the commit message editor with
+`nextCommitMessage`, `gg do publish` pre-selects `versionIncrement` and
+pre-fills the merge message with `mergeMessage`. git ignores the file.
+
+Whenever you change code in a repo, update `nextCommitMessage` of that repo
+so it describes everything that is not committed yet, not only your last
+step. Create the file if it does not exist. Change only the keys you mean
+to change and keep the rest — `commits` is written by gg, never by hand.
+
+```json
+{
+  "publishConfig": {
+    "nextCommitMessage": {
+      "firstLine": "Let the user pick the ticket repos",
+      "details": ["Ask whether repos are chosen automatically or manually"]
+    },
+    "mergeMessage": "Let the user pick the ticket repos manually",
+    "versionIncrement": "minor"
+  }
+}
+```
+
+- `firstLine`: imperative, at most 60 characters
+- `details`: optional, one entry per line
+- `mergeMessage`: the pull request title and the merge commit message of
+  the repo
+- `versionIncrement`: `patch`, `minor` or `major`, matching what the
+  ticket changes in the repo
+
+Set `mergeMessage` and `versionIncrement` as soon as the change of the repo
+is clear, and adjust them when it changes. Only `gg do commit` in the
+ticket folder reads `nextCommitMessage`; `gg one do commit` does not.
+
 ## Commit
 
 ```bash
@@ -144,7 +188,8 @@ Afterwards load the review-light skill and execute it.
 
 - Create a blog post for the current ticket
 - Update the index.jsonc and README.md
-- Create the configuration for gg do publish
+- Check `mergeMessage` and `versionIncrement` in `.gg/publish_config.json`
+  of each repo — `gg do publish` offers them as defaults
 - If the ticket was only planned, publish the project management repo only
 
 Ask the user to run the following command **manually**:

package/dna/dot-claude/skills/gg-commit/SKILL.md

@@ -22,7 +22,13 @@ whenever you are inside a repo.
 
 Look at `git status` and `git diff` in the repos of the ticket. Propose one
 imperative one-liner that says what the step does, not which files changed.
-Let the user confirm it.
+Start from `nextCommitMessage` in `.gg/publish_config.json` of each repo and
+update it there if it no longer matches the diff (see the development
+guide). Let the user confirm it.
+
+If the user runs `gg do commit` in the ticket folder without `-m`, the
+editor already shows each repo's `nextCommitMessage` — `-m` overrides it
+with one message for all repos.
 
 ## 3. Commit
 

package/dna/dot-claude/skills/gg-publish/SKILL.md

@@ -15,12 +15,15 @@ creates and pushes the version tags. It is interactive — the user runs it.
 
 ## 2. Configure the publish
 
-```bash
-gg do configure-publish
-```
+Write per repo into `.gg/publish_config.json` (see the development guide):
+
+- `mergeMessage` — the pull request title and merge commit message
+- `versionIncrement` — `patch`, `minor` or `major`, matching what the
+  ticket changed
 
-Check per repo that the version bump matches what the ticket changed and
-that `CHANGELOG.md` reads well.
+`gg do publish` pre-fills the merge message and pre-selects the version
+increment with these values. Never run `gg do configure-publish` yourself,
+it is interactive. Check that `CHANGELOG.md` reads well.
 
 ## 3. Let the user publish
 

package/dna/dot-claude/skills/gg-ticket/SKILL.md

@@ -52,10 +52,19 @@ Call it the "project management repo" whenever you write to the user, never
 
 ## 4. Choose the code repos
 
-Read the `index.md` of the candidate repos in `.ocean`. Tell the user which
-repos the ticket needs and what you roughly want to change in each one. If a
-part belongs to a domain that has no repo yet, say so and ask whether to
-create one. A ticket that is only planned may need no code repo at all.
+Ask the user how the code repos are chosen, as a single question with two
+options:
+
+- **automatically** — Claude finds the repos. Read the `index.md` of the
+  candidate repos in `.ocean`. Tell the user which repos the ticket needs
+  and what you roughly want to change in each one. If a part belongs to a
+  domain that has no repo yet, say so and ask whether to create one.
+- **manually** — the user names the repos. Ask in a second single input
+  field which repos the ticket needs, e.g. `repo1 repo2`. Check each name
+  against the repos in `.ocean` and ask again for names that do not exist
+  there. The entered list is the confirmation.
+
+A ticket that is only planned may need no code repo at all.
 
 After the confirmation:
 
@@ -81,6 +90,6 @@ repo alone.
 
 ## Important
 
-- Never add repos without confirmation. Rather too few than too many — more
-  can be added later.
+- Never add repos without confirmation — a manually entered list counts as
+  one. Rather too few than too many — more can be added later.
 - Do not start the implementation unasked.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@ggdna/dna-gg",
-  "version": "0.5.1",
+  "version": "0.7.0",
   "description": "DNA layer: gg CLI workflow guides, repo setup scripts and CI check",
   "keywords": [
     "dna",
@@ -28,12 +28,13 @@
     "access": "public"
   },
   "dependencies": {
+    "@ggdna/dna-github": "^0.0.2",
     "@ggdna/dna-index": "^0.1.2",
     "@ggdna/dna-install": "^0.1.2"
   },
   "devDependencies": {
-    "@types/node": "^26.6.2",
-    "prettier": "^3.9.8"
+    "@types/node": "^26.6.4",
+    "prettier": "^3.9.9"
   },
   "scripts": {
     "test:dart": "dart test",

package/dna/dot-github/workflows/quick_check.yaml

@@ -1,20 +0,0 @@
-# Runs the gg quick check on every push and pull request against main.
-
-name: Tests
-
-on:
-  push:
-    branches: ['main']
-  pull_request:
-    branches: ['main']
-
-jobs:
-  build:
-    name: dnaGitQuickCheckPipelineName
-    runs-on: ubuntu-latest
-
-    steps:
-      - uses: actions/checkout@v5
-
-      # Uses the Node.js preinstalled on the runner image.
-      - run: npx @tssuite/ggwsm one did commit