agentleFS
strikersamstrikersam/autonomous-ai-agency/.claude/skills/client-onboarding/SKILL.md

client-onboarding

Skill12 starsChanged 29 days agoMITView on GitHub

Onboard a new client/company onto this platform's real Agentic OS: create the company record, scan its websites/repos, auto-provision specialist agents, activate its 24x7 agency runtime, and know exactly which "OS" building blocks (memory, integrations, dashboard) already exist versus which are roadmap gaps.

  • Sends data out

8 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

1 rule. 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 8 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If the client asks for Gmail/Calendar/Drive automation, a notes vault, or manually-curated "skill packs": these are not implemented.

    line 129

“Common” is measured: kept by at least 6 in 10 of the 8 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Skill: client-onboarding

Fill these in

Fill inWhat it isExample
COMPANY_NAMEDisplay name for the new client company (required)Bright Studio
DOMAINPrimary domain for the company (required — CompanyCreateRequest.domain, models/company_graph.py:1966)brightstudio.com
WEBSITE_URLSOptional: public site(s) to scan for stack/systems detection["https://brightstudio.com"]
REPO_URLSOptional: repos to scan for code-level detection["https://github.com/brightstudio/app"]

Ownership is derived server-side from the caller's bearer token (_resolve_user_id(user) in backend/company_api.py) — there is no owner_id request field to fill in; do not invent one.

COMPANY_NAME and DOMAIN are required to create the company record. WEBSITE_URLS/REPO_URLS are optional — onboarding can run with website_urls empty and specialists can still be provisioned manually afterward.

What this maps to (real code, not a generic scaffold)

Giveaway-skill conceptThis repo's actual implementation
"Scaffold the memory layer / vault"CompanyGraphStore (services/company_graph_store.py:52) persists Company, CompanyGraph, Specialist, Workflow, KnowledgeItem, Connector, ApprovalPolicy (Mongo primary / SQLite fallback). There is no notes vault — knowledge lives in KnowledgeItem records (create_knowledge_item/search_knowledge, company_graph_store.py:337-378). Agent-runtime memory (separate concern) lives in agent/memory.py and agent/persistent_memory.py.
"Wire in the Google suite (Gmail/Calendar/Drive)"Does not exist yet. packages/auth/oauth.py only implements Google OAuth2 login (GOOGLE_CLIENT_ID/GOOGLE_CLIENT_SECRET, /api/auth/google/login) — no Gmail, Calendar, or Drive API integration. Do not tell a client this is available; if they need it, file it as a new Connector type (see models/company_graph.py) rather than faking it.
"Drop in requested skill packs (research, content, ops, data)"SpecialistService.provision_specialists_for_company (services/specialist.py) auto-provisions specialists from a ~33-family catalog (engineering, qa, security, devops, data, ml, frontend, backend, marketing, research, support, trading, portfolio, agile, ...) based on systems detected during scanning — not a manual pack install.
"Build the command-center dashboard"Already built: frontend/src/v5/screens/DashboardScreen.jsx (health/stats/activity/providers/tasks), AgentsScreen.jsx, CompanyScreen.jsx, SkillsScreen.jsx. Nothing to scaffold — the new company just needs to be onboarded into it.
"Brand it and write a README"Company display name + branding lives on the Company record itself (models/company_graph.py); no separate README artifact is generated per client.

Workflow

Step 1 — Create the company and kick off onboarding

Via API (backend/company_api.py):

# Create company record (router prefix is /api/company, see backend/company_api.py:96)
# body shape is CompanyCreateRequest (models/company_graph.py:1966): name + domain required
curl -X POST "$API_BASE/api/company" \
  -H "Authorization: Bearer $JWT" -H "Content-Type: application/json" \
  -d '{"name": "COMPANY_NAME", "domain": "brightstudio.com"}'

# Start the real onboarding pipeline for that company_id
curl -X POST "$API_BASE/api/company/{company_id}/onboarding/start" \
  -H "Authorization: Bearer $JWT" -H "Content-Type: application/json" \
  -d '{"website_urls": WEBSITE_URLS, "repo_urls": REPO_URLS}'

Or programmatically, from services/onboarding.py:

from services.onboarding import OnboardingService

service = OnboardingService()
progress = await service.start_onboarding(
    company_id="company_123",
    website_urls=["https://brightstudio.com"],
    repo_urls=[],
)

This runs the fixed 8-step pipeline (OnboardingService.ONBOARDING_STEPS): create_company → scan_websites → scan_repositories → detect_systems → provision_specialists → create_workflows → complete → activate_agency.

Step 2 — Poll progress

curl "$API_BASE/api/company/{company_id}/onboarding" -H "Authorization: Bearer $JWT"

Or via UI: frontend/src/v5/screens/OnboardingScreen.jsx renders the same OnboardingProgress model live.

Step 3 — Verify specialists were provisioned

curl "$API_BASE/api/company/{company_id}/specialists" -H "Authorization: Bearer $JWT"

Expect specialists matching the systems scan_websites/scan_repositories detected (e.g. a Next.js site + Stripe → frontend, ecommerce, payments specialists). If the list is empty, call POST /api/company/{company_id}/specialists to provision manually, or re-run onboarding with auto_provision_specialists=True.

Step 4 — Confirm the 24x7 agency runtime is live

The final activate_agency step calls CompanyAgency.activate_company() (services/company_agency.py) in the background. Confirm via the Doctor endpoint (GET /api/company/doctor/public) or AgentsScreen.jsx in the dashboard — specialists should show as running, not just provisioned.

Step 5 — Note real gaps instead of pretending they're solved

If the client asks for Gmail/Calendar/Drive automation, a notes vault, or manually-curated "skill packs": these are not implemented. Log them as a Connector/KnowledgeItem feature request (see models/company_graph.py) or a task in .claude/state/active-tasks.md rather than fabricating the integration. Do not add a dependency on an external community/product link — this platform is self-hosted (CLAUDE.md §1, Non-goals).

Acceptance checks

  • Company record created (GET /api/company/{company_id} returns 200)
  • Onboarding progress reaches complete then activate_agency
  • At least one specialist provisioned and visible in AgentsScreen.jsx
  • Doctor/health check for the company shows green
  • Any requested-but-unsupported integration (Gmail/Calendar/Drive, vault) is explicitly called out as unsupported, not silently skipped

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.