agentsclimarketplace

Obsidian resume

Skill dawnn07/obsidian-resume/skills/obsidian-resume

Use this skill whenever the user wants to generate a resume, CV, or job application materials from their Obsidian notes. Triggers on: "generate resume from Obsidian", "make my resume from my notes", "build CV from vault", "create Harvard resume", "find jobs matching my resume", "export resume as LaTeX", "resume from markdown notes". Also trigger when the user pastes Obsidian markdown and asks for a resume, cover letter, or job search. This skill reads exported vault JSON (from export_vault.py), generates a Harvard OCS-style LaTeX resume, and runs a live job search based on the user's profile.From its SKILL.md

Install
npx -y skills add dawnn07/obsidian-resume --skill obsidian-resume

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

9.7 KB, as published. Nobody here has run it

Obsidian → Harvard Resume + Job Search Skill

Overview

This skill does three things in sequence:

  1. Parse exported Obsidian notes (obsidian_export.json) into a structured profile
  2. Generate a Harvard OCS-style LaTeX resume (compiles to PDF with pdflatex)
  3. Search live job listings that match the user's skills and experience

Step 0 — Check for export file

Before anything else, ask the user:

"Have you run export_vault.py yet? If yes, paste the contents of obsidian_export.json here. If not, I can also work directly from pasted markdown notes."

If they have the JSON: parse it using the schema in references/export_schema.md.
If they paste raw markdown: extract profile data directly — see Parsing Rules below.


Step 1 — Parse Obsidian notes into a profile

Build a structured profile object with these fields:

profile:
  name: string
  email: string
  phone: string
  location: string
  linkedin: string (optional)
  github: string (optional)
  website: string (optional)
  summary: string (2–3 sentences, optional)

  education:
    - degree: string
      school: string
      location: string
      dates: string          # e.g. "Sep 2016 – Jun 2020"
      gpa: string (optional)
      honors: string (optional)
      relevant_courses: [string] (optional)

  experience:
    - title: string
      company: string
      location: string
      dates: string
      bullets: [string]      # strong action verbs, quantified impact

  projects:
    - name: string
      tech: string           # comma-separated stack
      dates: string (optional)
      bullets: [string]

  skills:
    technical: [string]
    languages: [string]
    tools: [string]
    soft: [string] (optional)

  awards: [string] (optional)
  publications: [string] (optional)
  volunteer: [string] (optional)

Parsing rules

  • Obsidian frontmatter keys (title:, company:, dates:, tags:) map directly to profile fields
  • Headings (## Experience, ## Skills) signal sections
  • Bullet points (-) under a job heading become bullets[]
  • Dataview fields (company:: Acme Corp) are treated like frontmatter
  • Tags #career, #resume, #job mark notes as relevant
  • If a field is missing, leave it out — do NOT invent data
  • Dates: preserve whatever format the user wrote; do not reformat

Step 2 — Generate Harvard OCS LaTeX resume

Read the full template in references/harvard_template.md before generating.

Harvard OCS rules (strict)

RuleDetail
Font10–12pt, Times New Roman or Latin Modern
Margins0.5–1 inch all sides
Section headersALL CAPS, bold, followed by a full-width horizontal rule
NameLarge (16–18pt), bold, centered at top
Contact lineSingle line, centered, below name
DatesRight-aligned on same line as job title/company
BulletsStart with strong past-tense action verb; quantify impact where possible
Length1 page for <10 years experience; 2 pages max
OrderEducation → Experience → Projects → Skills → Awards
No photosNo headshots or graphics
No colorsPure black and white (use \color{black} everywhere)

LaTeX output format

Output the complete .tex file — do not truncate. Structure:

\documentclass[10pt, letterpaper]{article}
% ... preamble (see references/harvard_template.md for full preamble)

\begin{document}

% ── HEADER ──────────────────────────────────────────────────
\begin{center}
  {\LARGE\textbf{FULL NAME}} \\[4pt]
  \small [email protected] \; $\cdot$ \; +1 (555) 000-0000 \; $\cdot$ \; City, Country \\
  \small \href{https://linkedin.com/in/handle}{linkedin.com/in/handle} \; $\cdot$ \; \href{https://github.com/handle}{github.com/handle}
\end{center}

\vspace{-6pt}

% ── EDUCATION ───────────────────────────────────────────────
\section*{EDUCATION}
\hrule
\vspace{4pt}

\noindent\textbf{University Name} \hfill City, Country \\
\textit{Degree, Major} \hfill \textit{Sep 20XX -- Jun 20XX} \\
GPA: X.X/4.0 \; | \; Honors: Dean's List

\vspace{6pt}

% ── EXPERIENCE ──────────────────────────────────────────────
\section*{EXPERIENCE}
\hrule
\vspace{4pt}

\noindent\textbf{Company Name} \hfill City, Country \\
\textit{Job Title} \hfill \textit{Jan 20XX -- Dec 20XX}
\begin{itemize}[leftmargin=1.5em, topsep=2pt, itemsep=0pt]
  \item Accomplished [X] by doing [Y], resulting in [Z\%] improvement
  \item Led team of N engineers to deliver feature used by X users
\end{itemize}

\vspace{6pt}

% ── PROJECTS ────────────────────────────────────────────────
\section*{PROJECTS}
\hrule
\vspace{4pt}

\noindent\textbf{Project Name} \textit{| Tech Stack} \hfill \textit{Month 20XX}
\begin{itemize}[leftmargin=1.5em, topsep=2pt, itemsep=0pt]
  \item Built [X] using [Y] that achieved [Z]
\end{itemize}

\vspace{6pt}

% ── SKILLS ──────────────────────────────────────────────────
\section*{SKILLS}
\hrule
\vspace{4pt}

\noindent\textbf{Languages:} Python, TypeScript, Go \\
\noindent\textbf{Frameworks:} React, FastAPI, Node.js \\
\noindent\textbf{Tools:} Docker, GitHub Actions, AWS, PostgreSQL

\end{document}

LaTeX special characters — escape these always

CharacterEscaped
&\&
%\%
$\$
#\#
_\_
{ }\{ \}
~\textasciitilde{}
^\textasciicircum{}
\\textbackslash{}

Vietnamese and accented characters work natively with \usepackage[T1]{fontenc} + \usepackage[utf8]{inputenc} — include both in preamble.

Bullet writing rules

Rewrite the user's raw notes into resume bullets:

  • Start with a past-tense action verb (Built, Led, Designed, Reduced, Launched, Automated...)
  • Include a metric whenever the notes hint at one (users, %, time saved, $ value, team size)
  • Keep each bullet to 1 line (≤90 chars) where possible
  • Do NOT invent metrics — if there are none, write a strong qualitative bullet

Step 3 — Live job search

After generating the resume, extract a search profile:

search_profile:
  top_skills: [3–5 most prominent technical skills]
  roles: [2–3 job titles that fit the experience level]
  location: [from profile, default "Ho Chi Minh City" or "Remote"]
  years_experience: [estimate from dates]
  seniority: [Junior / Mid / Senior / Lead]

Then use web search to find live job listings. Run these searches:

  1. "{role}" "{top_skill}" jobs {location} 2025
  2. "{role}" remote Vietnam 2025 site:linkedin.com OR site:vietnamworks.com OR site:topcv.vn
  3. "{role}" "{top_skill}" "{top_skill_2}" job opening

For each search, extract up to 3 real listings with:

  • Job title
  • Company
  • Location / Remote status
  • URL to apply
  • Why it matches (1 sentence referencing specific skills from the resume)

Present results as a table:

RoleCompanyLocationMatch reasonLink
............Apply

Then offer to write a tailored cover letter for any listing.


Step 4 — Cover letter (on request)

When the user picks a job, generate a cover letter:

  • Format: Plain text, 3 paragraphs, under 280 words
  • Para 1: Why this specific company/role (reference something real about the company)
  • Para 2: 2 strongest achievements from the resume that match the job
  • Para 3: Brief close + call to action
  • Tone: Confident, direct, not generic — avoid phrases like "I am writing to express my interest"

Output checklist

Before presenting the LaTeX to the user, verify:

  • All special characters are escaped
  • inputenc and fontenc packages included (for non-ASCII names)
  • Every \begin{} has a matching \end{}
  • Dates are right-aligned using \hfill
  • Section headers use \section*{} (no numbering) + \hrule
  • No invented data — only what came from the notes
  • File compiles with pdflatex resume.tex (mentally trace through)

Files in this skill

FilePurpose
SKILL.mdThis file — main instructions
scripts/export_vault.pyRun locally to export Obsidian vault → JSON
references/harvard_template.mdFull LaTeX preamble + section examples
references/export_schema.mdJSON schema for obsidian_export.json

Quick reference — how to use this skill

  1. Run python export_vault.py --vault ~/path/to/vault on your computer
  2. Paste the output JSON into Claude
  3. Claude generates your Harvard LaTeX resume
  4. Save the .tex file and compile: pdflatex resume.tex
  5. Claude searches live jobs matching your profile
  6. Pick a job → Claude writes a tailored cover letter

Keep looking

Skills are one crate of 608,475. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.