feat: initialize plugin marketplace with core plugins

- add marketplace.json, README, and .gitignore
- add git-tools plugin with commit-format skill
- add numencore-core plugin with skill-creator skill

.claude-plugin/marketplace.json

@@ -0,0 +1,22 @@
+{
+  "name": "numencore-toolkit",
+  "description": "Personal plugin marketplace — phatch",
+  "owner": {
+    "name": "Parley Hatch",
+    "email": "parley.hatch@gmail.com"
+  },
+  "plugins": [
+    {
+      "name": "numencore-core",
+      "description": "Core toolkit skills — skill authoring, project scaffolding",
+      "category": "development",
+      "source": "./plugins/numencore-core"
+    },
+    {
+      "name": "git-tools",
+      "description": "Git workflow skills — commit formatting, branch management",
+      "category": "development",
+      "source": "./plugins/git-tools"
+    }
+  ]
+}

.gitignore

@@ -0,0 +1,20 @@
+# OS
+.DS_Store
+Thumbs.db
+
+# Editor
+*.swp
+*.swo
+*~
+.vscode/
+.idea/
+
+# Temp
+*.tmp
+.claude/temp/
+
+# Local environment
+.claude/
+docs/
+CLAUDE.md
+

README.md

@@ -0,0 +1,73 @@
+# numencore-toolkit
+
+Personal Claude Code plugin marketplace. Self-hosted on Forgejo.
+
+## Install
+
+```bash
+git clone ssh://git@git.numencore.com:2222/phatch/numencore-toolkit.git
+claude plugin marketplace add ~/numencore-toolkit
+claude plugin install numencore-core@numencore-toolkit
+claude plugin install git-tools@numencore-toolkit
+```
+
+Verify with `claude plugin list` or `/reload-plugins` in an active session.
+
+## Plugin management
+
+```bash
+# List installed
+claude plugin list
+
+# Update marketplace (re-reads from disk)
+claude plugin marketplace update numencore-toolkit
+
+# Reinstall after editing skills (re-caches)
+claude plugin install <plugin>@numencore-toolkit
+
+# Disable/enable
+claude plugin disable <plugin>@numencore-toolkit
+claude plugin enable <plugin>@numencore-toolkit
+
+# Uninstall
+claude plugin uninstall <plugin>@numencore-toolkit
+```
+
+## Development
+
+After editing skills in the source directory, reinstall the plugin to update the cache:
+
+```bash
+claude plugin install <plugin>@numencore-toolkit
+```
+
+In an active session, run `/reload-plugins` to pick up changes.
+
+For faster iteration, bypass the cache entirely:
+
+```bash
+claude --plugin-dir ~/numencore-toolkit/plugins/<plugin-name>
+```
+
+### Structure
+
+```
+plugins/<plugin-name>/
+├── .claude-plugin/plugin.json
+└── skills/<skill-name>/SKILL.md
+```
+
+### Adding a new plugin
+
+1. Create `plugins/<name>/.claude-plugin/plugin.json`
+2. Add skills under `plugins/<name>/skills/<skill-name>/SKILL.md`
+3. Register in `.claude-plugin/marketplace.json`
+4. Run `claude plugin marketplace update numencore-toolkit`
+5. Run `claude plugin install <name>@numencore-toolkit`
+
+## Plugins
+
+| Plugin | Description |
+|--------|-------------|
+| `numencore-core` | Core toolkit — skill authoring, project scaffolding |
+| `git-tools` | Git workflow — commit formatting, branch management |

plugins/git-tools/.claude-plugin/plugin.json

@@ -0,0 +1,12 @@
+{
+  "name": "git-tools",
+  "version": "0.1.0",
+  "description": "Git workflow skills — commit formatting, branch management",
+  "author": {
+    "name": "Parley Hatch",
+    "email": "parley.hatch@gmail.com"
+  },
+  "repository": "ssh://git@git.numencore.com:2222/phatch/numencore-toolkit.git",
+  "license": "MIT",
+  "skills": "./skills/"
+}

plugins/git-tools/skills/commit-format/SKILL.md

@@ -0,0 +1,54 @@
+---
+name: commit-format
+description: Generate a conventional commit message from staged changes and commit after user approval. Use when the user wants to commit staged work.
+user-invocable: true
+argument-hint: "[type] [scope] [description]"
+allowed-tools: Bash(git *)
+---
+
+# Commit Format
+
+## Steps
+
+1. Read the staged change summary:
+   - `git diff --cached --stat` for file summary
+   - Do NOT read the full diff — infer intent from the stat summary and file names
+   - If nothing is staged, run `git status --short` and offer:
+     - Stage all changes
+     - Stage by file type or directory
+     - Let the user specify which files
+   - Wait for the user's choice and stage accordingly before continuing
+
+2. Determine commit metadata:
+   - If arguments were provided, use them as type/scope/description
+   - If no arguments, infer from the diff:
+     - **type**: `feat`, `fix`, `refactor`, `docs`, `test`, `chore`, `build`, `ci`, `perf`, `style`
+     - **scope**: infer from the primary area of change (module, component, file group) — omit if unclear
+     - **description**: one line summarizing what changed
+
+3. Build the commit message:
+   - Subject line: `type(scope): description` or `type: description` if no scope
+   - Subject line must be under 72 characters
+   - If the change warrants it, add a body separated by a blank line — short bullet points only
+   - No emojis, no attributions, no co-authored-by lines
+   - Body is for non-obvious context only — skip it if the subject says enough
+
+4. Present the message to the user in a fenced block:
+   ```
+   feat(parser): add support for nested expressions
+
+   - handle recursive descent for grouped terms
+   - update token precedence table
+   ```
+
+5. Wait for user response:
+   - If confirmed: run `git commit -m` with the message (use HEREDOC for multiline)
+   - If edits requested: revise and present again
+   - Never commit without explicit approval
+
+## Constraints
+
+- One sentence subject lines — not paragraphs
+- Prefer active voice, imperative mood ("add", "fix", "remove" — not "added", "fixes", "removing")
+- If the diff is large, focus the message on intent, not listing every file
+- When in doubt about type, ask the user rather than guess wrong

plugins/numencore-core/.claude-plugin/plugin.json

@@ -0,0 +1,12 @@
+{
+  "name": "numencore-core",
+  "version": "0.1.0",
+  "description": "Core toolkit skills — skill authoring, project scaffolding",
+  "author": {
+    "name": "Parley Hatch",
+    "email": "parley.hatch@gmail.com"
+  },
+  "repository": "ssh://git@git.numencore.com:2222/phatch/numencore-toolkit.git",
+  "license": "MIT",
+  "skills": "./skills/"
+}

plugins/numencore-core/skills/skill-creator/SKILL.md

@@ -0,0 +1,121 @@
+---
+name: skill-creator
+description: Build a new Claude Code skill from a user's description. Use when the user wants to create, author, or write a new skill or slash command.
+user-invocable: true
+argument-hint: "[what the skill should do]"
+allowed-tools: Read, Write, Bash(mkdir *), Glob, Grep
+---
+
+# Skill Creator
+
+You are building a Claude Code skill to the numencore-toolkit standard.
+
+## Phase 1: Understand
+
+Extract these six answers from the user's input. Ask for missing answers ONE AT A TIME. Do not proceed until all six are resolved.
+
+1. **What does it do?** — One sentence. If it takes more, the skill is too broad.
+2. **What does it take in?** — Arguments shape, or none.
+3. **What does it produce?** — File, diff, terminal output. Where does it go?
+4. **What tools does it need?** — Explicit list (Read, Write, Bash, Grep, Glob, etc).
+5. **What is the interaction shape?** — Execute immediately, or back-and-forth with user?
+6. **Does output have a specific format?** — If yes, define the shape.
+
+If the user is vague, push for clarity. If they say red, do not build blue.
+
+When all six are answered, play back a tight spec:
+
+```
+Skill: [name]
+Does: [one sentence]
+Input: [arguments or none]
+Output: [what and where]
+Tools: [list]
+Interaction: [shape]
+Format: [description or "none"]
+```
+
+Wait for explicit confirmation before proceeding.
+
+## Phase 2: Write
+
+Build the skill following these principles. Every line must earn its place.
+
+### Frontmatter rules
+
+- `name`: lowercase, hyphens, max 64 chars
+- `description`: pushy and specific — this is a routing mechanism, not documentation
+- `allowed-tools`: explicit list, never implicit
+- `argument-hint`: include if skill takes arguments
+- `disable-model-invocation`: set `true` for side-effect skills (deploy, send, delete)
+- Only include fields that have non-default values
+
+### Body rules
+
+- Every line is HOW or WHERE
+- Constraint-level WHY is allowed when it prevents bad judgment calls — but it earns its place
+- No filler, no commentary, no emojis, no preamble
+- Never repeat what frontmatter already declares
+- If output has a specific format, include ONE example of good output
+- Use `${CLAUDE_SKILL_DIR}` to reference bundled templates or scripts
+- Use `!`command`` preprocessor for dynamic context injection
+- Use `$ARGUMENTS`, `$0`, `$1` for argument substitution
+
+### Size rules
+
+- Target under 200 lines
+- If exceeding 200, move reference material to supporting files
+- Supporting files go in the skill directory alongside SKILL.md
+- Reference them with markdown links — they load on demand, not automatically
+
+### Structure rules
+
+- Self-contained — one job, no skill chaining
+- Instruction sections use `##` headers
+- Steps within sections are numbered
+- Constraints within steps are bulleted
+
+## Phase 3: Place
+
+1. Determine which plugin the skill belongs in
+2. Create the directory: `plugins/<plugin>/skills/<skill-name>/`
+3. Write `SKILL.md`
+4. Write any supporting files (templates, scripts, examples)
+5. Update `plugin.json` if this is a new plugin
+
+## Example Output
+
+A well-formed skill:
+
+```yaml
+---
+name: changelog
+description: Generate a changelog entry from staged git changes. Use when committing features, fixes, or breaking changes.
+user-invocable: true
+argument-hint: "[version]"
+allowed-tools: Bash(git *), Read, Write
+---
+
+# Changelog Generator
+
+## Steps
+
+1. Read staged diff: `git diff --cached --stat`
+2. Read recent changelog format from `CHANGELOG.md` if it exists
+3. Categorize changes into: Added, Changed, Fixed, Removed
+4. Write entry under `## [$0]` header at top of CHANGELOG.md
+
+Only include categories that have entries.
+
+## Format
+
+```markdown
+## [1.2.0]
+
+### Added
+- User authentication via OAuth2
+
+### Fixed
+- Connection timeout on large file uploads
+```
+```