Craigslist Monitor

Scrapes Craigslist NY service ads to find small owner-operated Staten Island, Brooklyn, and Bronx businesses needing more calls, ideal for lead generation.

Security Scan

The skill's behavior matches its stated purpose (Craigslist lead scraping) but contains environment hardcoding and dependency assumptions that don't fit a generic install and could cause unexpected imports or failures — review before installing.

Purpose & Capability

The code and SKILL.md both implement Craigslist scraping for Staten Island/Brooklyn/Bronx leads, which is coherent with the name/description. However, the skill depends on a local 'scrapling' package/venv and references local paths: SKILL.md uses ~/StudioBrain/... while monitor.py hardcodes a venv site-packages path (/Users/wlc-studio/...). That absolute path is not proportional to the stated purpose for a third-party user and will cause either import failures or unexpected imports if a different user's environment contains code at that path.

Instruction Scope

Instructions are narrowly scoped to fetching Craigslist search pages, optionally visiting ad pages to extract phone numbers, and appending results to a local MASTER_LEAD_LIST.md. There is no network communication to third-party endpoints beyond Craigslist. The code uses stealthy_headers and impersonate='chrome' in Fetcher.get (intended to avoid detection), which is a scraping tactic but not an obvious security violation by itself.

Install Mechanism

No install spec is provided, but monitor.py forcibly inserts a hardcoded site-packages path from a user-specific venv (/Users/wlc-studio/...). This is fragile and risky: it assumes a particular filesystem layout and will silently import modules from that path if present. There is no mechanism documented to install or verify the 'scrapling' dependency from a trusted source.

Credentials

The skill declares no required environment variables or credentials (appropriate for public Craigslist scraping), but it modifies sys.path to include an absolute venv path. That gives the script power to import arbitrary code from that path if present. The MASTER_LEAD_LIST path is under the user's home (~/StudioBrain/...), which is expected for local save but should be noted.

Persistence & Privilege

The skill is not always-enabled, does not request elevated privileges, and does not modify other skills or system-wide settings. Scheduling via cron is suggested in SKILL.md but is an optional action the user would take.

What to consider before installing

This skill appears to do what it says (scrape Craigslist and save leads) but has two red flags: a hardcoded venv site-packages path (/Users/wlc-studio/...) and an implicit dependency on a local 'scrapling' package that the repository doesn't install. Before installing or running: 1) inspect the 'scrapling' package and any local scrape.py referenced by SKILL.md — do not run if you don't trust its code. 2) Replace the hardcoded VENV_SITE with a user-relative path (e.g., using Path.home()) or install dependencies into a controlled virtualenv on your machine. 3) Run the script in a sandboxed environment (container or throwaway VM) first to confirm behavior. 4) Be aware the Fetcher uses 'stealthy' headers/impersonation — this may violate Craigslist's terms of service. If you need help checking the scrapling code or changing the import path, inspect monitor.py and the scrapling package files before use.

Craigslist Monitor Skill

Purpose

Scrape Craigslist NY services ads to find small owner-operated businesses in Staten Island, Brooklyn, and Bronx — these are Gracie AI Receptionist leads. Businesses advertising on Craigslist need more work, meaning more calls coming in, making them ideal Gracie prospects.

Location

~/StudioBrain/00_SYSTEM/skills/craigslist-monitor/

Usage

# Basic scan (all categories)
python3 ~/StudioBrain/00_SYSTEM/skills/craigslist-monitor/monitor.py

# Scan + save results to MASTER_LEAD_LIST.md
python3 ~/StudioBrain/00_SYSTEM/skills/craigslist-monitor/monitor.py --save

# Scan specific business type only
python3 ~/StudioBrain/00_SYSTEM/skills/craigslist-monitor/monitor.py --type plumber

# Also fetch individual ads to find phone numbers
python3 ~/StudioBrain/00_SYSTEM/skills/craigslist-monitor/monitor.py --save --enrich

What It Scrapes

Category	Search URL
Plumber	newyork.craigslist.org/search/sss?query=plumber&nearby=1
HVAC	newyork.craigslist.org/search/sss?query=hvac&nearby=1
Auto Repair	newyork.craigslist.org/search/sss?query=auto+repair&nearby=1
Dental	newyork.craigslist.org/search/sss?query=dental&nearby=1
Contractor	newyork.craigslist.org/search/sss?query=contractor&nearby=1

Output

• Console: Formatted list with business name, phone, location, URL
• --save: Appends markdown table to ~/StudioBrain/30_INTERNAL/WLC-Services/LEADS/MASTER_LEAD_LIST.md

Filters

Leads must be in: Staten Island, Brooklyn, or Bronx (by name or zip code). Leads with phone numbers are sorted first.

Dependencies

• Scrapling venv: ~/StudioBrain/00_SYSTEM/skills/scrapling/.venv/
• scrape.py: ~/StudioBrain/00_SYSTEM/skills/scrapling/scrape.py

Cron / Scheduled Use

Run weekly on Mondays to catch fresh Craigslist posts:

openclaw cron add --schedule "0 8 * * MON" --command "python3 ~/StudioBrain/00_SYSTEM/skills/craigslist-monitor/monitor.py --save"

Why Craigslist = Hot Leads

A business posting on Craigslist is actively seeking customers. That means:

• They have call volume (or want it)
• They're spending time/money on marketing
• They likely can't afford a full-time receptionist
• Gracie solves their exact problem