Files
settled-reach/whatsinagame/skill/references/profile-manifests.md
T
jpmschweitzerandClaude Opus 4.6 7158ae9cf5 feat(meta): add whatsinagame multi-agent team starter kit
Claude-native starter kit that bootstraps multi-agent team
infrastructure for any project. Clone once, install as a global
skill, run /kit-install in any project directory.

Includes:
- 3-tier profile system (minimal/standard/full: 3-12 agents)
- 16 agent archetype templates with personality spectrum
- 18 skill templates using domain-action naming convention
- Stakeholder persona panel for workshops and PR reviews
- SQLite ticketing DB with CLI tools (config-based DB paths)
- Decision tracking, sprint lifecycle, workshop orchestration
- Multi-git-host support (GitHub, Gitea, GitLab)
- /kit-update skill for syncing with source repo evolution
- Naming theme support for agent identity/flavor
- Smoke tests for all three profile tiers

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-18 02:22:30 +01:00

7.7 KiB

Profile Manifests

What each profile includes. The installer deploys these files adapted to the target project.

Minimal Profile (3 agents, 2 skills)

For solo developers or small projects that want structure without overhead.

Agents

Archetype Role Model
architect Technical architecture, feasibility, code review sonnet
project-manager Task tracking, prioritization, coordination haiku
qa-engineer Testing, bug investigation, quality assurance sonnet

Skills

Skill Trigger
git-commit /git-commit — structured conventional commits
skill-create /skill-create — guidance for writing new skills

Files Deployed

CLAUDE.md                           # Project instructions
.claude/settings.json               # Agent settings, model prefs
.claude/agents/architect.md         # Architect agent
.claude/agents/project-manager.md   # PM agent
.claude/agents/qa-engineer.md       # QA agent
.claude/skills/git-commit/SKILL.md   # Commit skill
.claude/skills/skill-create/SKILL.md # Skill creation guide
Makefile                            # Build/test/lint targets (if none exists)

Standard Profile (6 agents, 2 stakeholders, 8 skills)

For teams that want ticketing, decision tracking, and PR workflows.

Agents

Everything in minimal, plus:

Archetype Role Model
designer Systems design, UX, feature design sonnet
developer-backend Backend implementation, API design sonnet
content-author Written content, documentation, copy haiku

Stakeholder Personas

Persona Perspective
power-user Technical users who push limits, want efficiency
casual-user Non-technical users who want simplicity

Skills

Everything in minimal, plus:

Skill Trigger
ticket /ticket — manage project tickets in SQLite DB
pr-push /pr-push — push commits and create/update PRs
pr-review /pr-review — spawn parallel reviewers for branch diffs
worktree-update /worktree-update — sync worktree branches with main
sprint-retro /sprint-retro — run a sprint retrospective
release-notes /release-notes — generate release notes from commits

Files Deployed

Everything in minimal, plus:

TEAM.md                                 # Team roster and roles
decisions/README.md                     # Decision domain index
decisions/architecture.md               # Architecture decisions
decisions/scope.md                      # Scope decisions
decisions/process.md                    # Process decisions
.claude/agents/designer.md              # Designer agent
.claude/agents/developer-backend.md     # Backend developer agent
.claude/agents/content-author.md        # Content author agent
.claude/agents/power-user.md            # Power-user stakeholder
.claude/agents/casual-user.md           # Casual-user stakeholder
.claude/skills/ticket/SKILL.md          # Ticket skill
.claude/skills/pr-push/SKILL.md         # Push PR skill
.claude/skills/pr-review/SKILL.md       # Review PR skill
.claude/skills/pr-review/references/    # Review checklist references
.claude/skills/worktree-update/SKILL.md # Worktree sync skill
.claude/skills/sprint-retro/SKILL.md    # Retro skill
.claude/skills/release-notes/SKILL.md   # Release notes skill
.config/hooks/pre-commit                # Pre-commit hook
.config/hooks/commit-msg                # Commit message validation
db/schema.sql                           # Ticketing database schema
db/connectors/config.json               # Service endpoint config
db/connectors/ticket                    # Ticket CLI
db/connectors/sprint                    # Sprint CLI
db/connectors/sqlite_connector.py       # SQLite wrapper
db/connectors/sqlite-query              # SQL query helper
db/connectors/sqlite-exec               # SQL exec helper
db/connectors/sqlite-init               # DB initializer

Full Profile (12 agents, 4 stakeholders, 17 skills)

For ambitious projects that want the full orchestration toolkit.

Agents

Everything in standard, plus:

Archetype Role Model
consultant External perspective, architecture review, feasibility opus
visual-designer Art direction, UI consistency, visual coherence sonnet
audio-designer Sound design, audio pipelines, spatial audio sonnet
developer-frontend Frontend implementation, UI code, client systems sonnet
librarian Documentation, search indexing, knowledge management haiku
security-specialist Threat modeling, vulnerability assessment, secure code review sonnet

Stakeholder Personas

Everything in standard, plus:

Persona Perspective
product-advocate Champions the product vision, evaluates wow factor
marketer External messaging, positioning, what makes this sellable

Skills

Everything in standard, plus:

Skill Trigger
sprint-start /sprint-start — begin sprint work on a team branch
sprint-plan /sprint-plan — plan next sprint, write briefings
workshop-start /workshop-start — multi-agent design workshops
docs-search /docs-search — semantic search via Qdrant
team-standup /team-standup — quick status check across agents
health-check /health-check — verify infrastructure health
dep-audit /dep-audit — audit dependencies for issues
team-onboard /team-onboard — generate onboarding docs for new contributors
debt-scan /debt-scan — scan for technical debt

Files Deployed

Everything in standard, plus:

docs/DEVOPS.md                              # Build, test, lint, CI procedures
docs/briefings/architect.md                 # Per-agent briefings
docs/briefings/project-manager.md
docs/briefings/qa-engineer.md
docs/briefings/designer.md
docs/briefings/developer-backend.md
docs/briefings/content-author.md
docs/briefings/consultant.md
docs/briefings/visual-designer.md
docs/briefings/audio-designer.md
docs/briefings/developer-frontend.md
docs/briefings/librarian.md
docs/sprints/                               # Sprint directory scaffolding
docs/workshops/                             # Workshop directory scaffolding
.claude/agents/consultant.md                # Consultant agent
.claude/agents/visual-designer.md           # Visual designer agent
.claude/agents/audio-designer.md            # Audio designer agent
.claude/agents/developer-frontend.md        # Frontend developer agent
.claude/agents/librarian.md                 # Librarian agent
.claude/agents/product-advocate.md          # Product advocate stakeholder
.claude/agents/marketer.md                  # Marketer stakeholder
.claude/agents/security-specialist.md       # Security specialist agent
.claude/skills/sprint-start/SKILL.md        # Start sprint skill
.claude/skills/sprint-plan/SKILL.md         # Plan sprint skill
.claude/skills/sprint-plan/references/      # Sprint planning references
.claude/skills/workshop-start/SKILL.md      # Workshop skill
.claude/skills/docs-search/SKILL.md         # Semantic search skill
.claude/skills/team-standup/SKILL.md        # Standup skill
.claude/skills/health-check/SKILL.md        # Health check skill
.claude/skills/dep-audit/SKILL.md           # Dependency audit skill
.claude/skills/team-onboard/SKILL.md        # Onboarding skill
.claude/skills/debt-scan/SKILL.md           # Debt scan skill
db/connectors/qdrant_connector.py           # Qdrant + Ollama connector
db/connectors/qdrant-search                 # Semantic search CLI
db/connectors/qdrant-index                  # Document indexer CLI
db/connectors/qdrant-health                 # Qdrant health check
db/connectors/qdrant-count                  # Collection document count