init-cv

name	init-cv
description	Onboard a new master CV from a PDF or YAML file. Use when the user wants to set up or replace the master resume. Accepts a file path as the argument.
disable-model-invocation	true

/init-cv

Onboard a new master CV from a PDF or YAML file.

Usage

/init-cv <path-to-pdf-or-yaml>

Examples:

• /init-cv ~/Downloads/resume.pdf
• /init-cv ~/Documents/my_resume.yaml

Workflow

Follow these steps exactly in order:

Step 1 — Determine input type

Read the file at the provided path. Determine whether it is a PDF or a YAML file by its extension (.pdf vs .yaml / .yml).

If the file does not exist or the extension is unrecognised, report the error and stop.

Step 2 — Parse or validate

If PDF: Read the PDF content and extract all resume data. Produce a RenderCV-compliant YAML with the following structure:

cv:
  name: "Full Name"
  email: "[e-mail]"
  phone: "[phone]"
  location: "City, Country"
  website: "https://..."
  social_networks:
    - network: LinkedIn
      username: username
  sections:
    summary:
      - "..."
    experience:
      - company: "..."
        position: "..."
        location: "..."
        start_date: "YYYY-MM"
        end_date: "YYYY-MM"  # or "present"
        highlights:
          - "..."
    education:
      - institution: "..."
        area: "..."
        degree: "..."
        start_date: "YYYY"
        end_date: "YYYY"
    skills:
      - label: "Category"
        details: "Skill 1, Skill 2, Skill 3"

design:
  theme: classic
  typography:
    line_spacing: 0.8em
    font_size:
      body: 9.5pt
      headline: 9.5pt
      connections: 9.5pt
  page:
    show_footer: false
    show_top_note: false
  header:
    connections:
      phone_number_format: international
  section_titles:
    space_above: 0.8cm
  entries:
    date_and_location_width: 3cm
    highlights:
      space_above: 0.2cm
  sections:
    show_time_spans_in: []
    space_between_regular_entries: 0.2cm
  templates:
    experience_entry:
      main_column: "**COMPANY**, POSITION — LOCATION\nSUMMARY\nHIGHLIGHTS\n"
      date_and_location_column: "DATE"
    education_entry:
      main_column: "**INSTITUTION**, DEGREE in AREA — LOCATION\nSUMMARY\nHIGHLIGHTS"
      degree_column: null
      date_and_location_column: "DATE"
    normal_entry:
      main_column: "**NAME**\nSUMMARY\nHIGHLIGHTS\n"
      date_and_location_column: ""

Follow RenderCV entry type rules:

• Use company field for experience entries (ExperienceEntry)
• Use institution field for education entries (EducationEntry)
• Use name field for project entries (NormalEntry)
• Use label field for one-line entries (OneLineEntry)
• Do not mix entry types within a section
• Write cv.sections.summary in standard resume style with implied first person: no I, he, him, she, her, they, or similar pronouns.

Follow layout rules (see Layout Rules section below).

If YAML: Read the file and validate it conforms to RenderCV structure. Check that:

• Top-level keys include cv and optionally design and settings
• cv.name is present
• Each section uses a consistent entry type
• Dates follow YYYY-MM or YYYY format
• If cv.sections.summary exists, flag first-person or third-person pronouns in summary bullets and normalize them to standard resume style before saving.
• If settings.bold_keywords exists, remove it or set it to an empty list before saving. This repo does not use global keyword auto-bolding because it affects fixed sections like certifications.

Report any structural issues found. If the YAML is valid, proceed to Step 3.

Step 3 — Confirm overwrite if needed

Check whether templates/<name>_cv.yaml already exists (where <name> is the person's name, lowercased, spaces replaced by underscores).

If it does, ask the user to confirm before overwriting:

templates/<name>_cv.yaml already exists. Overwrite it? (yes/no)

Do not proceed until the user confirms.

Step 4 — Save the master CV

Save the parsed or validated YAML to templates/<name>_cv.yaml, where <name> is the person's name from the CV, lowercased, with spaces replaced by underscores (e.g., templates/david_alecrim_cv.yaml).

Step 5 — Verify it renders

Use the Makefile render target, which delegates to scripts/render.py to handle rendering cross-platform — output always lands in ./output/:

make render FILE=templates/<name>_cv.yaml

If the render succeeds, report the path to the generated PDF and confirm setup is complete.

If the render fails, show the error output and list what needs to be fixed. Do not delete the saved master_cv.yaml — let the user fix and re-run.

Layout Rules

These rules prevent common rendering issues in the classic theme and must be applied when producing YAML from a PDF or validating an existing YAML:

Experience entries:

• Use the summary field for a one-line company description (rendered as italic text under the position). Do not put the company description in highlights.
• Do not add "Stack:" or "Technologies:" bullets to highlights. Technology context belongs in the skills section.
• Keep each highlight to 1–2 lines of text. Overly long bullets break column alignment.

Education entries:

• Keep degree short: use abbreviations like BSc, MSc, Specialization, PhD.
• area holds the field of study (e.g., Computer Science, Software Engineering).
• Always include the design.templates.education_entry block to prevent degree column overflow and hyphenation. The degree_column: null moves DEGREE inline into the main column, and location is placed inline via the main_column template:

  design:
    theme: classic
    templates:
      education_entry:
        main_column: "**INSTITUTION**, DEGREE in AREA — LOCATION\nSUMMARY\nHIGHLIGHTS"
        degree_column: null
        date_and_location_column: "DATE"
  

Typography defaults:

• Keep the default compact typography block in this repo unless the user asks otherwise:

  design:
    theme: classic
    typography:
      line_spacing: 0.8em
      font_size:
        body: 9.5pt
        headline: 9.5pt
        connections: 9.5pt
    sections:
      show_time_spans_in: []
      space_between_regular_entries: 0.2cm
  

Project entries:

• Use plain text for name values. Do not use Markdown links in project names.
• Do not include start_date, end_date, or date fields. Project sections should not show a time window.
• Include the design.templates.normal_entry block:

  design:
    theme: classic
    templates:
      normal_entry:
        main_column: "**NAME**\nSUMMARY\nHIGHLIGHTS\n"
        date_and_location_column: ""
  

Skills entries:

• Keep each details value under ~80 characters. Long comma-separated lists wrap badly.
• Split into multiple label entries rather than cramming everything into one line.
• Avoid parenthetical expansions like AWS (EKS, RDS, S3, ...) inline — list the services separately if needed.

Summary entries:

• Keep summary bullets in resume style: concise noun-led or verb-led statements with no personal pronouns.
• Do not write the summary in first person (I ...) or third person (he ..., she ..., they ...).

Certifications:

• Use one OneLineEntry per certification. label is the full cert name; details is the date (required by RenderCV):

  - label: "AWS Certified Solutions Architect – Professional"
    details: "May 2024"
  - label: "Certified Kubernetes Administrator (CKA) – CNCF"
    details: "Aug 2023"
  

• Do not group by issuer. Full cert names as proper nouns are clearer and avoid awkward comma-separated lists.

Notes

• The output YAML under templates/ becomes the source of truth for all future /chameleon runs.
• If rendercv is not installed, tell the user to run make install-tools before retrying Step 5.