This is a submission for the Hermes Agent Challenge

What I Built

Job hunting is tedious. You find a posting, you open your CV, you squint at both for 20 minutes trying to figure out if you're actually a good fit, then you spend another hour writing a cover letter that probably sounds like every other cover letter.

I'm a full-stack and AI engineer currently on the market. I thought: what if I could just send a job URL to my phone and get back a scored analysis and a tailored cover letter while I'm on the bus?

That's what I built with Hermes Agent.

Job Match Agent is a custom Hermes skill that:

1. Reads your CV from a PDF
2. Fetches a job description from a URL (or accepts pasted text)
3. Scores your fit out of 100 with a detailed breakdown across four criteria
4. Writes a tailored cover letter using your actual CV content and the job's own keywords
5. Delivers everything back to you on WhatsApp

The whole thing runs on my Windows machine. I talk to it from my phone. One message in, full analysis back.

Demo

Flow: Type "Job match this: [URL]" on WhatsApp → Hermes fetches the job, reads your CV, scores fit, writes cover letter → full response back on your phone in ~2 minutes.

Code

GitHub: hermes-job-match-agent

My Tech Stack

• Hermes Agent — orchestration, skill system, memory, WhatsApp bridge
• Python 3.12 — core scripting
• pdfplumber — PDF CV extraction (pypdf fallback)
• httpx — job URL fetching with browser User-Agent spoofing
• beautifulsoup4 — job description extraction with prioritised selectors for LinkedIn, Greenhouse, Lever, Workday
• WhatsApp self-chat bridge — built into Hermes Agent gateway

How I Used Hermes Agent

This is where it gets interesting. I didn't write a script that does these steps mechanically. I wrote a skill a markdown file that teaches Hermes Agent what to do and Hermes handles all the planning and reasoning itself.

When I send a job URL, here's what Hermes actually does autonomously:

• Loads the job-match skill from its skills directory
• Calls read_cv.py to extract my CV text from the PDF on disk
• Calls fetch_job.py to scrape the job description from the URL
• Reasons through a structured scoring rubric across four dimensions
• Writes a cover letter that mirrors the job's language naturally
• Sends the full response back over WhatsApp

That's real multi-step agentic behavior; planning, tool execution, reasoning, output, not a hardcoded pipeline. Hermes decides the order of operations, handles failures gracefully, and formats the final response for WhatsApp automatically.

The memory system is what makes the WhatsApp UX smooth. I told Hermes my CV path once:

"Please remember my CV is always at: C:\Users...\Christian_Nnorom.pdf"

Hermes saved it. Now from my phone I just send:

"Job match this: [URL]"

No repeating myself. No attaching files. Hermes already knows.

The Architecture

The two Python scripts are focused tools. All the intelligence the planning, scoring rubric, and cover letter logic, lives in the skill's instructions, which Hermes reads and follows.

The Skill

Hermes skills are markdown files with YAML frontmatter. No plugin registration, no framework boilerplate. Drop a SKILL.md into the skills folder and Hermes picks it up automatically.

---
name: job-match
description: "Analyze a job posting against a CV/resume to score fit, explain reasoning, and generate a tailored cover letter."
version: 1.0.0
author: Christian Nnorom
platforms: [linux, macos, windows]
metadata:
  hermes:
    tags: [Jobs, Career, CV, Resume, Cover Letter, Recruitment]
---

The skill defines the four-step workflow in plain English, with scoring criteria, output format templates, and fallback handling. Hermes reads this and executes it — including deciding when to run which script and how to handle edge cases like a URL that requires JavaScript rendering.

Real Results

I tested it against two live job postings on the same day I built it.

Role 1 — KTP Associate, AI & Computer Vision (UWE Bristol / Foster + Freeman)

Overall Score: 64/100

Required Skills: 20/40 — Strong ML/CV experience but no PhD (mandatory)
Experience Level: 12/20 — 3+ years industry experience, MSc distinction
Domain Match: 14/20 — Applied AI/CV work relevant, forensic context less so
Nice-to-haves: 18/20 — Docker, AWS, MLOps, full-stack all present

Verdict: Apply. Lead with production AI pipeline experience. Address the
PhD gap directly by highlighting MSc distinction and immediate readiness.

Role 2 — AI Platform Engineer, KTP Associate (Aston University / Modular Data Ltd)

Overall Score: 84/100

Required Skills: 32/40 — Strong match on cloud/microservices, MLOps, Python,
  ethical AI (SHAP/LIME). Gaps: knowledge graphs, Go/Java.
Experience Level: 18/20 — MSc Distinction + 3+ years production AI systems
Domain Match: 16/20 — Modular AI platform work aligns directly
Nice-to-haves: 18/20 — Open-source, AWS cert, multi-cloud, ethical AI all present

Verdict: Strong fit. Emphasise modular platform delivery and MLOps expertise.

Two roles, two honest scores — 64 and 84. The agent called out the PhD gap on the first role rather than inflating the score. That's exactly what you want from a tool like this.

Each analysis came with a full cover letter — 248 words, keywords matched from the job description, no generic openers.

The Code

read_cv.py — tries pdfplumber first, falls back to pypdf, prints extracted text with a word/line count header. Fails loudly with actionable errors if extraction fails.

fetch_job.py — uses httpx with a real browser User-Agent, then beautifulsoup4 to extract the job container using a prioritised list of selectors covering LinkedIn, Greenhouse, Lever, Workday, and generic semantic tags. Falls back gracefully with instructions to paste text directly if the page requires JavaScript.

pip install pdfplumber httpx beautifulsoup4

Full code at: github.com/nnoromiv/job-match

What I Learned About Hermes Agent

Skills are surprisingly powerful. Writing the workflow in markdown felt almost too simple, but it works because Hermes actually reads and follows the instructions rather than vaguely gesturing at them.

The memory system is the difference between a demo and a real tool. Storing the CV path once and never repeating it makes the WhatsApp UX feel like a product, not a hack.

Honest agentic output requires honest prompting. The scoring rubric explicitly says "be honest in the gap analysis a useful score requires accurate assessment." Without that, agents flatter. With it, you get a 64 that's actually useful.

What's Next

• GPU - Running hermes on a gpu to quicken process.
• Batch mode — analyse 10 jobs at once, ranked by fit score
• Application tracker — Hermes logs each analysis to its Kanban board automatically
• Interview prep — after scoring, generate likely interview questions based on the gaps

The foundation is solid. Hermes Agent handles the orchestration; extending the skill is all it takes.

Try It Yourself

1. Install Hermes Agent
2. Clone github.com/nnoromiv/job-match
3. Copy the job-match folder into your Hermes skills directory
4. Run pip install pdfplumber httpx beautifulsoup4
5. Tell Hermes your CV path once via memory
6. Send any job URL and let it work

Built in one day for the Hermes Agent Challenge. Running on a Windows VPS, reachable from WhatsApp.