prime-sync

$npx mdskill add avibebuilder/claude-prime/prime-sync

Syncs Claude config between prime repo and target projects based on user input

  • Prompts users to specify which project(s) to sync or selects known projects if available.
  • Deploys changes to specified projects in push mode, imports changes from specified projects in pull mode.
  • Decides the operating mode by checking for a `VERSION` file or `.claude/.prime-version` file in the current working directory.
  • Updates state files with project paths and sync times after successful operations.

SKILL.md

.github/skills/prime-syncView on GitHub ↗
---
name: prime-sync
description: "Syncs Claude config between prime repo and target projects. Trigger on 'sync', 'prime-sync', 'push config', 'pull config', 'update prime', 'sync claude config'. Push mode deploys to targets; pull mode imports changes back."
argument-hint: "[<target-project-path>] (push mode only, optional in pull mode)"
disable-model-invocation: true
---

Ultrathink.

## Mode Detection

Detect operating mode before anything else:

1. `VERSION` file exists in CWD → **push mode** (CWD is the prime repo)
2. `.claude/.prime-version` exists in CWD → **pull mode** (CWD is a target project)
3. Neither detected → ask user which mode and for the required path

## Push Mode — Resolve Target

Source = CWD (prime repo). Target = resolved project path.

**State file:** `.claude/prime-projects.json`

Tracks known target paths, last sync time, and version. Shape:
```json
{ "projects": [{ "path": "/absolute/path", "lastSynced": "2025-12-30T10:30:00Z", "version": "1.4.2" }] }
```
Create as `{ "projects": [] }` if missing.

**Resolution flow:**
1. **Argument provided** → Use that path, then record it after a successful sync
2. **No argument, known projects exist** → Ask the user which project(s) to sync, including an **All projects** option
3. **No argument, no known projects** → Ask the user for a path

**State file management:**
- Create `.claude/` only if you need to persist the state file
- Update the synced project's time/version only after a successful sync
- Keep one entry per absolute path

## Pull Mode — Clone Prime

Source = cloned prime repo. Target = CWD.

**Flow:**
1. Read current version from `.claude/.prime-version` in CWD
2. Generate timestamp: `date +%Y%m%d%H%M%S`
3. Clone prime repo: `git clone https://github.com/avibebuilder/claude-prime.git /tmp/claude-prime-sync-<timestamp>/`
4. Set source = `/tmp/claude-prime-sync-<timestamp>/`, target = CWD
5. Continue to shared process below
6. Clean up `/tmp/claude-prime-sync-<timestamp>/` after sync completes (success or failure)

No state file in pull mode — the target project is self-contained.

## Process

### 1. Validate Target
- Check that the target path exists, is a directory, and is not the same location as the source repo
- Treat a valid target as a project root where writing `.claude/` is appropriate
- Read versions from target (`.claude/.prime-version`) and prime (`VERSION`)

If the path fails those checks, abort and explain why.

### 2. Parallel Analysis

**Change Detection**:
- Detect relevant changes in prime's `.claude/` based on the target's recorded version state
- If the target has no recorded version, or already matches the current prime version, sync only uncommitted prime changes
- If the target is on an older prime version, diff from that version tag to `HEAD` and include uncommitted changes; warn if the tag is missing
- Categorize changes into commands, agents, hooks, settings, skills, and starter-skills

**Stack Detection** (only if skills or starter-skills changed): check target for stack indicators relevant to each changed skill/starter.

**File Comparison**: compare each changed file with target. Detect: NEW, UPDATE, IDENTICAL, CONFLICT.

### 3. Build Sync Plan
Aggregate results and present:
```
CHANGED (will sync): ...
IDENTICAL (skip): ...
CONFLICTS (will ask): ...
IRRELEVANT (skip, wrong stack): ...
```

**GATE**: User approves sync plan.

### 4. Handle Conflicts
For each conflict, show the relevant diff and ask the user how to proceed:
- **Overwrite** — Replace the target copy with the prime version
- **Skip** — Leave the target copy unchanged
- **Merge** — Surface the full diff and let the user choose which parts to keep before writing anything

### 5. Execute Sync
1. Create `.claude/` in target if needed
2. Copy approved files (skills as entire folders)
3. Sync approved starter-skills to `.claude/starter-skills/` in target
4. Update `.prime-version` with prime's VERSION

### 6. Report
```
Sync complete! (prime vX.X.X)
Updated: ...
Skipped: ...
Version: X.X.X → written to .prime-version
```

Push mode only: update state file (`lastSynced` timestamp + `version`).
Pull mode only: clean up `/tmp/claude-prime-sync-*` clone directory.

## Gotchas

- **`prime-sync` stays in prime** — do not copy this skill into target projects.
- **Push and pull keep different state** — state file is push-only; pull mode treats the target as self-contained.
- **Version history can be incomplete** — if the expected tag is missing, warn and continue with that uncertainty explicit.

## Constraints

- NEVER modify the target project's source code
- NEVER sync without user approval at the plan gate
- In push mode, preserve one state entry per target path and update it only after success
- In pull mode, always clean up the temporary clone, even on failure

## Target Path

<target-path>$ARGUMENTS</target-path>

More from avibebuilder/claude-prime

SkillDescription
askAnswer questions about code, architecture, and technical decisions — no implementation. Trigger on questions asking 'why', 'what does this do', 'what is the purpose of', 'explain', 'what's the difference', 'compare', or 'what are the tradeoffs' — even when referencing specific files, code snippets, or inline code. The key signal is the user wants to UNDERSTAND something, not change it. Do NOT trigger for requests to build, fix, plan, review, research, or add/modify code.
backend-fastapi-pythonUse this skill for any Python backend work in this project: building FastAPI endpoints, writing service functions, defining Pydantic/SQLModel schemas, running Alembic migrations, or debugging 422 errors. Essential for authentication and authorization patterns — setting up get_current_user, is_superuser checks, admin-only guards, role-based access, and dependency injection chains like Depends(). Also covers middleware, background tasks, async SQLAlchemy sessions, ORM relationship loading, and request/response design. Activate whenever the question involves Python API code, FastAPI patterns, or backend architecture in this codebase. Not for frontend, Docker, CI/CD, or infrastructure.
create-docUse when the user wants to save knowledge as a file so others don't have to rediscover it — \"turn this into a doc\", \"write this up\", \"document how X works\", \"we figured this out and want to capture it\", \"nobody should have to figure this out again\". Covers any request to create or update durable written artifacts: onboarding guides, runbooks, ADRs, API docs, architecture notes, postmortems, changelogs, setup guides. The trigger: user wants knowledge captured in a file for future reference, not just a conversation. Do NOT use when still making decisions (→ give-plan), just asking for explanation without a file (→ ask), or writing code (→ cook).
discussBrainstorms and debates approaches, then drives toward an actionable decision. Use whenever someone needs a thinking partner for a decision they're facing: 'discuss', 'debate', 'brainstorm', 'weigh options', 'tradeoffs', 'should I do X or Y', 'help me decide', 'I'm torn between', 'sanity check my thinking', or 'what do you think about'. The user must be asking for help reasoning through a choice — not asking to build, fix, evaluate, plan, or modify something (even if the topic involves this skill itself). Picks the right decision lens, surfaces tradeoffs and blind spots, pushes back when reasoning is genuinely weak, and never implements.
docs-seekerFetch up-to-date documentation for any library, framework, API, or service into context. Use when the user wants to look up API references, check function signatures or required fields, find feature-specific docs, or verify how an external tool actually works. Triggers for queries about third-party libraries like Stripe, SQLAlchemy, Tailwind, FastAPI, shadcn, Drizzle, Hono, Better Auth — any time the answer lives in official docs rather than in the project codebase. Use this instead of guessing from trained knowledge, which is stale.
frontend-developmentUse this skill for ANY work involving React, Next.js, TypeScript, or Tailwind in the browser layer. This includes building components and pages, but equally covers debugging and fixing frontend issues: CSS/Tailwind classes not applying, form validation behavior, hydration mismatches between server and client renders, styling bugs, layout shifts, and rendering problems. Also use for refactoring components (e.g., splitting Server vs Client Components), data fetching patterns, state management, bundle optimization, and frontend tooling. If the problem involves what users see or interact with in a web browser — whether building, fixing, or refactoring — use this skill. Not for backend APIs, databases, infrastructure, or DevOps.
give-planUse when the user wants a written, reviewable plan or spec produced before coding starts. Triggers on: mapping out changes without implementing, thinking through risks of upgrades or migrations, evaluating approaches before committing to one, writing specs for team review, phasing work into stages, or any request that explicitly defers coding ('don't implement yet', 'before we build'). The distinguishing signal is that the user wants a plan artifact — not implementation, not a conversational answer. MUST activates inside Claude's native plan mode to have a better planning behavior.
media-processorSpecialized visual and multimedia processing tools. Use this skill whenever a task involves complex visual content — UI mockups, dense screenshots, design images, charts, artwork — where precise details like spacing, hex colors, font sizes, and component hierarchy need to be extracted accurately. Also use for: reviewing or auditing existing UI against designs, comparing screenshots for visual regressions, transcribing audio/video, extracting data from PDFs with complex layouts, and generating images. Trigger whenever the user wants to implement from a design, review or compare UI screenshots, analyze visual details precisely, describe artwork or aesthetic content, or process any media file (audio, video, PDF).
optimus-primeGenerates a Claude Code configuration tailored to a specific project. Use whenever the user wants to prime a project, set up claude for a repo, bootstrap claude config, or re-prime/refresh an already-primed project (for example after `/prime-sync` pulled new starter content). Triggers on 'prime', 'prime this project', 'optimus-prime', 're-prime', 'refresh claude config', 'regenerate CLAUDE.md', 'set up claude for this repo'. Deeply analyzes the real codebase and builds project-specific skills, rules, and CLAUDE.md — not generic boilerplate. For ongoing config health checks and proposal review, use `self-evolve` instead.
self-evolve>