Getting Started

Getting Started

Alhazen has two modes of operation. Choose the one that fits your workflow:

Mode	Use When	Guide
A. Claude Code (local)	Development, exploration, one-off analysis	This page
B. OpenClaw (persistent)	Overnight foraging, Telegram notifications, always-on	Deployment

This page covers Mode A: running Alhazen interactively with Claude Code on your local machine.

Prerequisites

Before you begin, ensure you have:

1. Claude Code installed and configured
2. Docker for running TypeDB
3. uv for Python dependency management

Installation

1. Clone the Repository

git clone https://github.com/gullyburns/skillful-alhazen
cd skillful-alhazen

2. Install Python Dependencies

# Install uv if you haven't already
curl -LsSf https://astral.sh/uv/install.sh | sh

# Install all dependencies
uv sync --all-extras

3. Start TypeDB

make db-start

This starts TypeDB on port 1729. The database schema initializes automatically on first use.

4. Open Claude Code

claude

Claude reads the skill files from .claude/skills/ and is ready to work.

Development Workflow: A → B → C

As your usage matures, you can progress to more persistent deployments:

Stage	Setup	Best For
A. Local Claude Code	This page	Skill development, ad hoc analysis
B. OpenClaw on Mac Mini	Deployment: Mac Mini Native + Deployment	Persistent foraging, Telegram triage
C. OpenClaw on VPS	Deployment (VPS section)	Always-on, cloud-hosted

Your First Session

Example 1: Job Hunting

Add and analyze a job posting:

You: I found a job at Anthropic:
     https://boards.greenhouse.io/anthropic/jobs/123456

Claude: I'll ingest and analyze this posting...

[Fetches the job description]
[Extracts company info, requirements, qualifications]
[Compares against your skill profile]
[Creates fit analysis with gap identification]

## Analysis: Research Engineer at Anthropic

**Your Fit: 82%**

| Requirement | Level | You | Match |
|-------------|-------|-----|-------|
| Python | Required | Strong | ✓ |
| ML Systems | Required | Strong | ✓ |
| Distributed Systems | Required | Some | △ |
| Rust | Preferred | None | ✗ |

Key Gap: Distributed systems is required; recommend DDIA + MIT 6.824.

Want me to set a priority level for this position?

Example 2: Rare Disease Investigation

Investigate a rare disease from its MONDO ID:

You: Investigate NGLY1 deficiency for me

Claude: I'll search for the MONDO ID and build a knowledge graph...

[Searches Monarch Initiative for "NGLY1 deficiency"]
[Finds MONDO:0800044]
[Initializes disease entity in TypeDB]
[Ingests HPO phenotypes, causal genes, MONDO hierarchy]
[Queries ClinicalTrials.gov for active trials]

## NGLY1 Deficiency (MONDO:0800044)

**Causal Gene:** NGLY1 (HGNC:17646) — total-loss mechanism
**Phenotypes:** 167 HPO terms; core: alacrima, global dev delay, seizures
**Hierarchy:** Congenital disorder of deglycosylation → rare disease
**Clinical Trials:** 2 active trials

What sensemaking would you like? I can analyze the mechanism,
therapeutic landscape, or differential diagnosis.

Example 3: Build Your Skill Profile (for Job Hunting)

Before analyzing jobs, set up your skill profile for gap analysis:

You: My skills are:
     - Python: strong (5 years production)
     - Machine Learning: strong (built models at CZI)
     - Kubernetes: some (deployed but not configured)
     - Rust: learning (started last month)

Claude: [Creates skill profile with levels]

Added 4 skills to your profile. I'll use these for fit analysis.

Dashboard

The project includes a Next.js dashboard for visualizing your job search:

make dashboard-dev   # starts on http://localhost:3000

Views:

• Pipeline Board (/jobhunt) — Kanban view by application status
• Position Detail (/jobhunt/position/{id}) — Requirements, notes, fit score
• Skills Matrix — Gap analysis across positions

Environment Variables

Variable	Default	Description
TYPEDB_HOST	localhost	TypeDB server hostname
TYPEDB_PORT	1729	TypeDB server port
TYPEDB_DATABASE	alhazen_notebook	Database name

Next Steps

• Jobhunt: See Skills: Jobhunt for the automated forager, full sensemaking workflow, and OpenClaw cron setup
• Rare Disease: See Skills: Rare Disease for the full 5-phase curation workflow
• Deploy persistently: Follow Deployment to set up OpenClaw for overnight foraging
• Architecture: Read Design Concepts to understand the curation philosophy and TypeDB role