@erclx/aitk 3.10.0 → 3.11.1
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/claude/.claude-plugin/plugin.json +1 -1
- package/claude/skills/claude-address-review/SKILL.md +10 -0
- package/claude/skills/claude-markdown-propose/REQUIREMENT.md +48 -0
- package/claude/skills/claude-markdown-propose/SKILL.md +118 -0
- package/claude/skills/claude-markdown-propose/references/format.md +107 -0
- package/claude/skills/claude-orchestrate/references/orchestrator-poll.md +4 -1
- package/claude/skills/claude-orchestrate/scripts/poll.sh +59 -13
- package/claude/skills/claude-pr-review/SKILL.md +4 -2
- package/claude/skills/claude-teach/SKILL.md +1 -1
- package/claude/skills/claude-worktree/SKILL.md +1 -1
- package/claude/skills/create-snippet/SKILL.md +1 -1
- package/claude/skills/git-branch/SKILL.md +1 -1
- package/claude/skills/git-commit/SKILL.md +1 -1
- package/claude/skills/git-issue/SKILL.md +2 -2
- package/claude/skills/git-pr/SKILL.md +5 -5
- package/claude/skills/git-split/SKILL.md +2 -2
- package/claude/skills/git-stage/SKILL.md +1 -1
- package/docs/agents/audits.md +2 -2
- package/docs/agents/records.md +1 -1
- package/docs/agents/teach.md +1 -1
- package/docs/ai-workflow.md +7 -6
- package/package.json +1 -1
- package/scripts/core/check-skill-paths.sh +1 -2
- package/scripts/core/verify.sh +0 -5
- package/src/audits/catalog.ts +20 -0
- package/src/claude/skills-audit.ts +7 -0
- package/src/commands/claude.ts +20 -6
- package/src/commands/comments.ts +6 -0
- package/src/commands/context.ts +32 -7
- package/src/commands/gov.ts +2 -2
- package/src/commands/markdown.ts +18 -2
- package/src/context/audit.ts +16 -0
- package/src/gov/test-order.ts +31 -7
- package/src/markdown/files.ts +10 -0
- package/src/records/backup.ts +1 -0
- package/src/records/validate.ts +1 -3
- package/{claude/skills/git-split/references → standards}/branch.md +0 -1
- package/{claude/skills/git-commit/references → standards}/commit.md +0 -1
- package/{claude/skills/claude-teach/references → standards}/glossary.md +0 -1
- package/standards/index.md +6 -0
- package/standards/{bundled/issue.md → issue.md} +0 -1
- package/standards/{bundled/pr.md → pr.md} +0 -1
- package/standards/{bundled/snippets.md → snippets.md} +0 -1
- package/claude/skills/claude-worktree/references/branch.md +0 -60
- package/claude/skills/create-snippet/references/snippets.md +0 -78
- package/claude/skills/git-branch/references/branch.md +0 -60
- package/claude/skills/git-issue/references/issue.md +0 -95
- package/claude/skills/git-pr/references/branch.md +0 -60
- package/claude/skills/git-pr/references/pr.md +0 -139
- package/claude/skills/git-split/references/pr.md +0 -139
- package/claude/skills/git-stage/references/commit.md +0 -73
- package/scripts/core/regen-skill-references.sh +0 -27
- package/standards/bundled/branch.md +0 -60
- package/standards/bundled/commit.md +0 -73
- package/standards/bundled/glossary.md +0 -76
|
@@ -1,73 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
title: Commit reference
|
|
3
|
-
description: Commit message format and type conventions
|
|
4
|
-
consumers: git-commit, git-stage
|
|
5
|
-
---
|
|
6
|
-
|
|
7
|
-
# Commit message reference
|
|
8
|
-
|
|
9
|
-
## Scope
|
|
10
|
-
|
|
11
|
-
Governs a git commit message: subject structure, the type and scope vocabulary, and the body.
|
|
12
|
-
|
|
13
|
-
Does not govern:
|
|
14
|
-
|
|
15
|
-
- Branch naming, which shares the type vocabulary: `branch.md`
|
|
16
|
-
- Pull request title and body, which share the subject form: `pr.md`
|
|
17
|
-
- Whether a phase label or a semver tag may appear in a subject: `versioning.md`
|
|
18
|
-
|
|
19
|
-
## Format
|
|
20
|
-
|
|
21
|
-
- Structure: `<type>(<scope>): <subject>`
|
|
22
|
-
- Casing: lowercase for `<type>`, `<scope>`, and first word of `<subject>`
|
|
23
|
-
- Subject: 72 characters maximum, no trailing period
|
|
24
|
-
|
|
25
|
-
## Types
|
|
26
|
-
|
|
27
|
-
- `feat`: new feature or capability
|
|
28
|
-
- `fix`: bug fix
|
|
29
|
-
- `refactor`: structural changes (not a fix or feature)
|
|
30
|
-
- `docs`: documentation only (README)
|
|
31
|
-
- `chore`: maintenance tasks (deps, tooling, configs)
|
|
32
|
-
- `perf`: performance improvements
|
|
33
|
-
- `test`: add or modify tests
|
|
34
|
-
- `style`: code formatting (whitespace, semicolons)
|
|
35
|
-
- `build`: build system changes (webpack, npm scripts)
|
|
36
|
-
- `ci`: CI/CD pipeline changes (GitHub Actions)
|
|
37
|
-
- `revert`: revert a previous commit
|
|
38
|
-
|
|
39
|
-
## Scope vocabulary
|
|
40
|
-
|
|
41
|
-
- Single lowercase word representing a system component
|
|
42
|
-
- Prefer single word
|
|
43
|
-
- Use kebab-case only when two words are genuinely needed for specificity
|
|
44
|
-
- Do not use specific filenames as scopes
|
|
45
|
-
- Do not use a scope that duplicates the type
|
|
46
|
-
- Write scopes for release readability. They surface in `changelogithub` release notes.
|
|
47
|
-
|
|
48
|
-
## Subject
|
|
49
|
-
|
|
50
|
-
- Use imperative mood (`add` not `added`)
|
|
51
|
-
- Describe the actual technical change, not that something changed
|
|
52
|
-
- Do not use vague verbs (`improve`, `refine`, `enhance`)
|
|
53
|
-
- Do not repeat the scope in the subject line
|
|
54
|
-
- Use single quotes if quoting
|
|
55
|
-
- No backslash escaping or internal double quotes
|
|
56
|
-
- No conversational filler or introductory phrases
|
|
57
|
-
|
|
58
|
-
## Examples
|
|
59
|
-
|
|
60
|
-
### Correct
|
|
61
|
-
|
|
62
|
-
```plaintext
|
|
63
|
-
feat(api): add retry logic for failed webhooks # specific verb + clear change
|
|
64
|
-
fix(auth): update 'UserSession' validation logic # scoped + imperative + single quotes
|
|
65
|
-
```
|
|
66
|
-
|
|
67
|
-
### Incorrect
|
|
68
|
-
|
|
69
|
-
```plaintext
|
|
70
|
-
fix(user-auth): Fixed the redirect loop. # wrong casing + period + multi-word scope
|
|
71
|
-
docs(docs): update the readme. # duplicate scope + period
|
|
72
|
-
docs(api): improve documentation # vague verb + lacks specificity
|
|
73
|
-
```
|
|
@@ -1,27 +0,0 @@
|
|
|
1
|
-
#!/usr/bin/env bash
|
|
2
|
-
set -e
|
|
3
|
-
set -o pipefail
|
|
4
|
-
|
|
5
|
-
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
|
|
6
|
-
PROJECT_ROOT="${PROJECT_ROOT:-$(cd "$SCRIPT_DIR/../.." && pwd)}"
|
|
7
|
-
|
|
8
|
-
source "$PROJECT_ROOT/scripts/lib/frontmatter.sh"
|
|
9
|
-
|
|
10
|
-
BUNDLED_DIR="$PROJECT_ROOT/standards/bundled"
|
|
11
|
-
SKILLS_DIR="$PROJECT_ROOT/claude/skills"
|
|
12
|
-
|
|
13
|
-
[ -d "$BUNDLED_DIR" ] || exit 0
|
|
14
|
-
|
|
15
|
-
while IFS= read -r file; do
|
|
16
|
-
filename="$(basename "$file")"
|
|
17
|
-
consumers="$(read_frontmatter_field "$file" "consumers")"
|
|
18
|
-
[ -z "$consumers" ] && continue
|
|
19
|
-
IFS=',' read -ra skills <<<"$consumers"
|
|
20
|
-
for skill in "${skills[@]}"; do
|
|
21
|
-
skill="${skill// /}"
|
|
22
|
-
[ -z "$skill" ] && continue
|
|
23
|
-
dest_dir="$SKILLS_DIR/$skill/references"
|
|
24
|
-
mkdir -p "$dest_dir"
|
|
25
|
-
cp "$file" "$dest_dir/$filename"
|
|
26
|
-
done
|
|
27
|
-
done < <(find "$BUNDLED_DIR" -type f -name "*.md" | sort)
|
|
@@ -1,60 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
title: Branch reference
|
|
3
|
-
description: Branch naming format and type conventions
|
|
4
|
-
consumers: git-branch, git-split, git-pr, claude-worktree
|
|
5
|
-
---
|
|
6
|
-
|
|
7
|
-
# Branch reference
|
|
8
|
-
|
|
9
|
-
## Scope
|
|
10
|
-
|
|
11
|
-
Governs a git branch name: its structure, its length, and the type vocabulary it draws from.
|
|
12
|
-
|
|
13
|
-
Does not govern:
|
|
14
|
-
|
|
15
|
-
- Commit subject format, which shares the type vocabulary: `commit.md`
|
|
16
|
-
- Pull request title and body: `pr.md`
|
|
17
|
-
- Whether a phase label may appear in a branch name: `versioning.md`
|
|
18
|
-
- Deriving a slug from a branch name for use in an output filename: `slug.md`
|
|
19
|
-
|
|
20
|
-
## Format
|
|
21
|
-
|
|
22
|
-
- Structure: `<type>/<description>` or `<type>/<ticket>-<description>`
|
|
23
|
-
- Length: 50 characters maximum
|
|
24
|
-
- Casing: kebab-case only, no underscores or camelCase
|
|
25
|
-
- Description: 2 words maximum, 3 only when genuinely needed for specificity
|
|
26
|
-
- Capture the core change, not the commit message verbatim
|
|
27
|
-
- For branches with multiple commits, use the unifying concern as the description.
|
|
28
|
-
- Do not duplicate type in description (e.g., `feat/feature-login`)
|
|
29
|
-
|
|
30
|
-
## Types
|
|
31
|
-
|
|
32
|
-
- `feat`: new feature or capability
|
|
33
|
-
- `fix`: bug fix
|
|
34
|
-
- `refactor`: structural changes (not a fix or feature)
|
|
35
|
-
- `docs`: documentation only (README)
|
|
36
|
-
- `chore`: maintenance tasks (deps, tooling, configs)
|
|
37
|
-
- `perf`: performance improvements
|
|
38
|
-
- `test`: add or modify tests
|
|
39
|
-
- `style`: code formatting (whitespace, semicolons)
|
|
40
|
-
- `build`: build system changes (webpack, npm scripts)
|
|
41
|
-
- `ci`: CI/CD pipeline changes (GitHub Actions)
|
|
42
|
-
- `revert`: revert a previous commit
|
|
43
|
-
|
|
44
|
-
## Examples
|
|
45
|
-
|
|
46
|
-
### Correct
|
|
47
|
-
|
|
48
|
-
```plaintext
|
|
49
|
-
feat/jwt-expiration # clear feature scope
|
|
50
|
-
fix/AUTH-123-connection-pool # includes ticket ID
|
|
51
|
-
refactor/remove-deprecated-endpoints # clear refactor intent
|
|
52
|
-
```
|
|
53
|
-
|
|
54
|
-
### Incorrect
|
|
55
|
-
|
|
56
|
-
```plaintext
|
|
57
|
-
feature/auth_stuff # wrong type + underscore
|
|
58
|
-
feat/feature-add-login # duplicates type in description
|
|
59
|
-
fix/DB-456-fix-the-database-connection-pool-memory-leak # exceeds 50 chars + verbatim message
|
|
60
|
-
```
|
|
@@ -1,73 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
title: Commit reference
|
|
3
|
-
description: Commit message format and type conventions
|
|
4
|
-
consumers: git-commit, git-stage
|
|
5
|
-
---
|
|
6
|
-
|
|
7
|
-
# Commit message reference
|
|
8
|
-
|
|
9
|
-
## Scope
|
|
10
|
-
|
|
11
|
-
Governs a git commit message: subject structure, the type and scope vocabulary, and the body.
|
|
12
|
-
|
|
13
|
-
Does not govern:
|
|
14
|
-
|
|
15
|
-
- Branch naming, which shares the type vocabulary: `branch.md`
|
|
16
|
-
- Pull request title and body, which share the subject form: `pr.md`
|
|
17
|
-
- Whether a phase label or a semver tag may appear in a subject: `versioning.md`
|
|
18
|
-
|
|
19
|
-
## Format
|
|
20
|
-
|
|
21
|
-
- Structure: `<type>(<scope>): <subject>`
|
|
22
|
-
- Casing: lowercase for `<type>`, `<scope>`, and first word of `<subject>`
|
|
23
|
-
- Subject: 72 characters maximum, no trailing period
|
|
24
|
-
|
|
25
|
-
## Types
|
|
26
|
-
|
|
27
|
-
- `feat`: new feature or capability
|
|
28
|
-
- `fix`: bug fix
|
|
29
|
-
- `refactor`: structural changes (not a fix or feature)
|
|
30
|
-
- `docs`: documentation only (README)
|
|
31
|
-
- `chore`: maintenance tasks (deps, tooling, configs)
|
|
32
|
-
- `perf`: performance improvements
|
|
33
|
-
- `test`: add or modify tests
|
|
34
|
-
- `style`: code formatting (whitespace, semicolons)
|
|
35
|
-
- `build`: build system changes (webpack, npm scripts)
|
|
36
|
-
- `ci`: CI/CD pipeline changes (GitHub Actions)
|
|
37
|
-
- `revert`: revert a previous commit
|
|
38
|
-
|
|
39
|
-
## Scope vocabulary
|
|
40
|
-
|
|
41
|
-
- Single lowercase word representing a system component
|
|
42
|
-
- Prefer single word
|
|
43
|
-
- Use kebab-case only when two words are genuinely needed for specificity
|
|
44
|
-
- Do not use specific filenames as scopes
|
|
45
|
-
- Do not use a scope that duplicates the type
|
|
46
|
-
- Write scopes for release readability. They surface in `changelogithub` release notes.
|
|
47
|
-
|
|
48
|
-
## Subject
|
|
49
|
-
|
|
50
|
-
- Use imperative mood (`add` not `added`)
|
|
51
|
-
- Describe the actual technical change, not that something changed
|
|
52
|
-
- Do not use vague verbs (`improve`, `refine`, `enhance`)
|
|
53
|
-
- Do not repeat the scope in the subject line
|
|
54
|
-
- Use single quotes if quoting
|
|
55
|
-
- No backslash escaping or internal double quotes
|
|
56
|
-
- No conversational filler or introductory phrases
|
|
57
|
-
|
|
58
|
-
## Examples
|
|
59
|
-
|
|
60
|
-
### Correct
|
|
61
|
-
|
|
62
|
-
```plaintext
|
|
63
|
-
feat(api): add retry logic for failed webhooks # specific verb + clear change
|
|
64
|
-
fix(auth): update 'UserSession' validation logic # scoped + imperative + single quotes
|
|
65
|
-
```
|
|
66
|
-
|
|
67
|
-
### Incorrect
|
|
68
|
-
|
|
69
|
-
```plaintext
|
|
70
|
-
fix(user-auth): Fixed the redirect loop. # wrong casing + period + multi-word scope
|
|
71
|
-
docs(docs): update the readme. # duplicate scope + period
|
|
72
|
-
docs(api): improve documentation # vague verb + lacks specificity
|
|
73
|
-
```
|
|
@@ -1,76 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
title: Glossary reference
|
|
3
|
-
description: Frontmatter, entry shape, ordering, and the rules deciding which terms a glossary carries
|
|
4
|
-
consumers: claude-teach
|
|
5
|
-
---
|
|
6
|
-
|
|
7
|
-
# Glossary reference
|
|
8
|
-
|
|
9
|
-
Applies to a glossary, the file holding one entry per term a body of material defines. It changes whenever the material names a concept a reader cannot look up yet, and it is revised in place rather than appended to.
|
|
10
|
-
|
|
11
|
-
## Scope
|
|
12
|
-
|
|
13
|
-
Governs a glossary at `.claude/teach/<nn>-<topic>/GLOSSARY.md` and at whatever path a surface fixes for one it holds: its frontmatter, entry shape, ordering, grouping, and the rules deciding which terms it carries.
|
|
14
|
-
|
|
15
|
-
Does not govern:
|
|
16
|
-
|
|
17
|
-
- The folder a learning workspace lays out around its glossary, and the other files in it: `teach.md`
|
|
18
|
-
- Which surface a glossary moves to once it leaves the material that produced it, which belongs to the surface driving that move
|
|
19
|
-
- Voice, rhythm, and sentence construction: the `write-human` skill
|
|
20
|
-
- Headings, punctuation, word choice, and file references: `markdown.md`
|
|
21
|
-
|
|
22
|
-
## What a working glossary looks like
|
|
23
|
-
|
|
24
|
-
A glossary works when a reader who meets a term in the material settles it here without opening the page that introduced it:
|
|
25
|
-
|
|
26
|
-
- Which word does this material use for the concept, and which words does it deliberately not use?
|
|
27
|
-
- What does the term mean, stated without leaning on the term itself?
|
|
28
|
-
- Where does the term appear, so a reader can see it used rather than only defined?
|
|
29
|
-
- Does every entry carry a term the material actually uses?
|
|
30
|
-
|
|
31
|
-
A glossary failing these is non-conforming even when it satisfies every shape rule below.
|
|
32
|
-
|
|
33
|
-
## Frontmatter
|
|
34
|
-
|
|
35
|
-
- `title` (required): names the material the terms come from, in sentence case
|
|
36
|
-
- `description` (required): one line naming what a reader gets from the entries
|
|
37
|
-
|
|
38
|
-
## Entries
|
|
39
|
-
|
|
40
|
-
- Write one entry per term, as a single bullet.
|
|
41
|
-
- Lead the bullet with the term as a bolded span, then the definition in one or two sentences.
|
|
42
|
-
- Define the term without using it. A definition that spends the term explains nothing to the reader who arrived not knowing it.
|
|
43
|
-
- Name where the term first appears, so a reader can reach one use of it in context.
|
|
44
|
-
- Keep an entry to the meaning. Worked detail belongs on the page that teaches the term.
|
|
45
|
-
- Sort entries alphabetically, so a reader who knows only the word finds it without reading the file.
|
|
46
|
-
|
|
47
|
-
## Which terms it carries
|
|
48
|
-
|
|
49
|
-
- Add a term once the material has used it, never ahead of that. A glossary front-loaded with terms nothing has introduced is a syllabus rather than a reference.
|
|
50
|
-
- Pick one word per concept and use that word everywhere. A glossary carrying two words for one thing hands the reader a choice it exists to remove.
|
|
51
|
-
- List each rejected synonym as an alias to avoid inside the entry that won, so a reader arriving with the wrong word lands on the right one.
|
|
52
|
-
- Use the glossary's own terms inside other definitions. A definition reaching for a synonym of a term defined two entries down teaches the reader a word the material does not use.
|
|
53
|
-
- Revise an entry the material has moved under rather than adding a second one narrating the change.
|
|
54
|
-
|
|
55
|
-
## Grouping
|
|
56
|
-
|
|
57
|
-
- Keep a short glossary as one alphabetical list under the title. Grouping a handful of entries costs a heading per category and saves no lookup.
|
|
58
|
-
- Group a glossary long enough that one list stops helping under `##` headings by category, sorted alphabetically within each. Roughly two screens of entries is the signal.
|
|
59
|
-
- Name each category so a reader picks it from the term alone. A category a reader cannot predict makes the grouping a second thing to search.
|
|
60
|
-
- State a departure from any rule above in the file itself, naming what it departs from and why. A glossary serving no single body of material is the case that produces one, since a term drawn from everywhere has no first appearance to name.
|
|
61
|
-
|
|
62
|
-
## Template
|
|
63
|
-
|
|
64
|
-
```markdown
|
|
65
|
-
---
|
|
66
|
-
title: <Material the terms come from>
|
|
67
|
-
description: <one line naming what a reader gets from these entries>
|
|
68
|
-
---
|
|
69
|
-
|
|
70
|
-
# <Material the terms come from>
|
|
71
|
-
|
|
72
|
-
<One line on which material these terms come from and when the file changes.>
|
|
73
|
-
|
|
74
|
-
- **<Term>**: <the meaning in one or two sentences, written without using the term>. Avoid <rejected synonym>. First appears in `<page or lesson>`.
|
|
75
|
-
- **<Term>**: <the meaning in one or two sentences, written without using the term>. First appears in `<page or lesson>`.
|
|
76
|
-
```
|