version: "1.0.1" name: setup description: Initialize, upgrade, or migrate the claude-harness plugin in a project. Use for first-time setup, plugin installation, version migration, memory directory configuration, legacy file cleanup, and project harness initialization. disable-model-invocation: true

Setup - Initialize or Upgrade Claude Harness

Initialize or upgrade claude-harness in the current project directory.

Auto-Detection

This command automatically detects what needs to be done:

• Fresh install: Creates v3.0 structure from scratch
• v2.x upgrade: Migrates existing files to v3.0 memory architecture
• Cleanup: Removes legacy command copies from .claude/commands/ (commands are now served from plugin cache)

Note: Commands and hooks are served from the plugin cache. setup.sh only initializes project-level state (memory directories, CLAUDE.md, .gitignore, migrations).

Phase 0: Run setup.sh (Handles ALL Cases)

setup.sh handles fresh installs, v2.x migrations, and cleanup in a single script. Always run it.

Steps:

1. Find the plugin root path from the session context (look for "Plugin Root:" in the session start context)
2. Run: bash {plugin-root}/setup.sh

• Fresh install: Creates v3.0 structure, .gitignore patterns, CLAUDE.md
• v2.x migration: Detects legacy files, migrates to v3.0 structure, then creates missing files
• Cleanup: Removes stale command copies from .claude/commands/ and legacy hooks
• Existing project files are NEVER overwritten (skipped automatically)
• .gitignore patterns are added if missing

1. Report what was created vs what was skipped
2. Skip Phase 3 -- setup.sh handles it

Fallback: If the plugin root path is not available in the session context, fall through to Phase 1 -> Phase 4 (manual setup).

Plugin updates: Run claude plugin update claude-harness to update the plugin itself.

Phase 3: Update Project .gitignore (FALLBACK -- only if setup.sh unavailable)

CRITICAL: You MUST update the project's .gitignore to exclude harness ephemeral files. This prevents uncommitted file clutter after /checkpoint.

Execute these steps:

1. Read the current .gitignore file (create if missing)

1. Check if .claude-harness/sessions/ pattern exists in the file

1. If the pattern is NOT present, append these lines to .gitignore:

``` # Claude Harness - Ephemeral/Per-Session State .claude-harness/sessions/ .claude-harness/memory/compaction-backups/

# Claude Code - Local settings .claude/settings.local.json ```

1. Use the Edit tool to append these patterns to .gitignore

1. Report: "Updated .gitignore with harness ephemeral patterns"

DO NOT SKIP THIS PHASE - it is required for proper harness operation.

File Schemas

Canonical schemas are defined in the plugin's schemas/ directory (JSON Schema format). Key state files:

Memory knowledge layers (memory/decisions/, failures/, successes/, patterns/, rules/) are an OKF v0.1 bundle -- markdown concept files with YAML frontmatter, not JSON. Their conformance rules are documented in schemas/okf-memory.md and checked by scripts/check-okf.py; setup.sh migrates legacy JSON memory files to the bundle automatically (stamp-gated). Semantic memory (memory/semantic/*.json) and runtime state stay JSON with v3 schemas as created by setup.sh.

After Setup

Report:

• What was done (fresh install / migration / version update)
• Files created or migrated
• Current plugin version
• Next steps:

1. Run /claude-harness:start to compile context and sync GitHub
2. For new projects: Run /claude-harness:prd-breakdown @./prd.md to analyze PRD and extract features
3. Use /claude-harness:flow "description" for end-to-end automated workflow (recommended)
4. Use /claude-harness:flow --no-merge "description" for step-by-step control
5. Use /claude-harness:flow --fix feature-XXX "bug" to create bug fixes