PJAlgoMaster

ApplyPilot — Development skill for Claude Code

Development community

Autonomous job-application agent - scrapes four boards, tailors an ATS resume per JD with Claude, submits through your own logged-in browser, DMs recruiters and cold-emails HR, then tracks replies and.

How to install ApplyPilot

This entry records only its repository, not the path inside it, so there is no exact command to give. Open PJAlgoMaster/ApplyPilot and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What ApplyPilot does

Alternatives in Development

Resumeskills — 20 specialized skills for resume optimization, ATS analysis, interview prep, and career transitions 160 ★
London Property Hunt Public — Automated London flat/room hunt powered by Claude Code + Claude in Chrome + Gmail MCP 85 ★
Replies — /replies — Sync, Classify & Triage Campaign Replies 66 ★

README

ApplyPilot

An agent that runs the job hunt for you. It scrapes security, cloud/devops AND software/full-stack roles, rewrites your own two base resumes into one JD-tailored ATS resume per job with Claude, submits the application through your own logged-in browser session, DMs a recruiter on LinkedIn, and cold-emails HR with the PDF attached — then reads the replies and nudges whoever went quiet. Capped at 80 applications/day — a deliberate mixture of the four families below, not whichever one the boards answer with first.

FastAPI + HTMX + Tailwind (both from CDN) + SQLite + Playwright. No Node, no build step.

Generation runs through the official **Claude Code CLI** in a hidden subprocess (`claude -p`), not the Anthropic Python SDK: the router in front of the models fingerprints HTTP clients and rejects the SDK with `401 unauthorized_client_error`, while it accepts the CLI. So `claude` must be installed and authenticated (`npm i -g @anthropic-ai/claude-code`, then `claude` once to log in).

1. Install

python -m venv .venv                # use Python 3.12 or 3.13 — see note below
.venv\Scripts\activate              # Windows;  source .venv/bin/activate elsewhere
pip install -r requirements.txt
playwright install chromium

**Python version:** everything except scraping runs on 3.14. `python-jobspy` and `pandas` have no 3.14 wheels yet, so they build from source and fail without a C compiler. Use 3.12/3.13 and the whole app works, `/scrape` included.

**weasyprint on Windows** needs the GTK runtime installed system-wide once: → Windows install. Without it, PDF rendering raises on import.

2. Add your source resumes

Every generated resume is written *from your own two resumes* — nothing about the candidate is hardcoded. Drop both files here:

source_resumes/cdac_resume.pdf        # structured / academic / credential-forward
source_resumes/personal_resume.pdf    # modern / dense / impact-forward

`ai_engine`