Initialize Enterprise Researcher: configure mTLS certificates, create the .enterprise-researcher data directory, and validate connectivity.
From enterprise-researchernpx claudepluginhub rsolipuram/enterprise-researcher/setupInitializes or resumes project setup via interactive Q&A, creating conductor/ artifacts for product definition, guidelines, tech stack, workflow, and style guides.
/setupDetects ghost Claude plugin installations by checking cache, registry, and temp files on macOS/Linux/Windows; cleans up on user confirmation.
/setupChecks local Codex CLI readiness, prompts to install if unavailable via npm, and optionally toggles stop-time review gate.
/setupGuides enterprise admins through Claude Office add-in setup for Vertex AI, Bedrock, or custom gateway; provisions credentials and generates deployable manifest.xml.
/setupConfigures claude-hud as Claude Code statusline by providing ~/.claude/settings.json config, build instructions, manual setup, and troubleshooting steps.
/setupRuns interactive setup wizard: detects AI providers (Codex, Gemini, Ollama, etc.), RTK token optimizer, system tools; displays status table; offers to install/configure missing items.
You are setting up the Enterprise Researcher knowledge base for the first time. Walk the user through configuration step by step.
Check if .enterprise-researcher/config.yaml already exists in the current project directory. If it does, inform the user and ask if they want to reconfigure.
Create the following directory structure in the project root:
.enterprise-researcher/
├── config.yaml
├── registry.yaml
├── specs/
├── index/
│ ├── domains/
│ └── services/
└── .gitignore
Ask the user for their mTLS certificate paths using AskUserQuestion:
/etc/pki/tls/client.crt)/etc/pki/tls/client.key)/etc/pki/tls/ca-bundle.crt)All three are required. Validate that the files exist at the given paths using Bash.
Write .enterprise-researcher/config.yaml with the mTLS paths and default settings:
mtls:
client_cert: <user-provided-path>
client_key: <user-provided-path>
ca_cert: <user-provided-path>
sync:
stale_after_hours: 24
timeout_seconds: 30
index:
data_dir: .enterprise-researcher
Write .enterprise-researcher/registry.yaml with an empty services list and commented examples:
# Enterprise Researcher — Service Registry
# Add your internal API services here
#
# Each service needs:
# name: Unique identifier (kebab-case)
# spec_url: Full URL to the OpenAPI JSON spec
# domain: Business domain grouping
# team: Owning team name
# description: Brief purpose description
# tags: List of searchable tags
services: []
Write .enterprise-researcher/.gitignore to exclude sensitive files:
# mTLS certificates should never be committed
*.key
*.pem
# Specs are large but can be committed (they're not secret)
# Uncomment the next line to exclude them:
# specs/
Ask the user if they'd like to test mTLS connectivity. If yes, ask for a test URL (any internal API URL) and run:
curl -s -o /dev/null -w "%{http_code}" --cert <cert> --key <key> --cacert <ca> <test_url>
Report the result.
Tell the user:
/add-service to register services one at a time.enterprise-researcher/registry.yaml directly for bulk registration/import-services to bulk-import from a file or URL/sync to fetch specs and build the knowledge index