AgentKit SEO
VitaeGraphGitHub

Methodology playbook

Build a Private Markdown Career Knowledge Graph

Five-minute guide

Start with the decisions that matter.

This short guide gives you the useful path first. The complete reference continues below, already open for deeper work and source review.

01

Use this when

  • A single compact context file is too flat for detailed education, thesis, course, project, or publication records.
  • An agent needs to create or maintain hierarchical Markdown records before using selected context in downstream tasks.
  • A local graph needs duplicate-ID, link, schema, or index validation.
02

What the method does

  1. Plan the graph from available material before writing records.
  2. Create focused Markdown nodes for education, projects, experience, certifications, awards, and publications.
  3. Validate record IDs and internal links, then generate deterministic local indexes as rebuildable artifacts.

Search and answer coverage

Questions this vitaegraph guide answers

Use the complete reference below for the constraints, examples, evidence rules, and implementation details behind each answer.

Complete referenceVitaeGraph playbook

VitaeGraph is the deep-record layer for AgentKit SEO. It complements the compact personal career context file with private, local Markdown records and rebuildable indexes.

Section: Overview

Use VitaeGraph when career material is too detailed for one compact context file. The compact file remains the fast agent handoff: identity, current goals, core facts, constraints, and high-level proof. VitaeGraph stores deeper records for education, thesis work, university courses, repository-backed projects, experience, certifications, awards, and publications.

The graph is local-first. Source records are Markdown files the user can read and edit. Generated JSON indexes are rebuildable artifacts.

Section: When to use it

Use VitaeGraph when:

  • A degree contains thesis work, courses, publications, or projects that deserve their own records.
  • A project needs more detail than a short context-file bullet can carry.
  • A GitHub repository URL is available and the agent should inspect public repo context before summarizing the project.
  • An agent needs selected deep context instead of the whole career history.
  • Record IDs, internal links, and generated indexes need validation.

Do not use VitaeGraph as a replacement for the compact career context file. Use it as the deeper companion when the task needs more structure.

Section: Default private workspace

The default private workspace convention is:

~/.agentkit-seo/<name-surname>-career-context.md
~/.agentkit-seo/vitaegraph/

The context file and VitaeGraph folder are separate artifacts. They may link to each other, but each remains useful on its own.

Section: Basic workflow

  1. Inspect the supplied career material and plan the records before writing.
  2. Create hierarchical Markdown records for the material that exists.
  3. Keep university courses under the related education folder when they belong to a degree.
  4. Keep thesis work under the related education folder when it belongs to a degree.
  5. Enrich repository-backed projects with public GitHub context when a repository URL is supplied.
  6. Validate the graph for duplicate IDs, internal links, and supported field values.
  7. Rebuild indexes from the Markdown source records.

Section: CLI commands

Initialize a graph:

npx agentkit-seo graph init --root ~/.agentkit-seo/vitaegraph

Validate a graph:

npx agentkit-seo graph validate --root ~/.agentkit-seo/vitaegraph

Generate deterministic indexes:

npx agentkit-seo graph index --root ~/.agentkit-seo/vitaegraph

Section: Privacy and limits

VitaeGraph does not create a backend, cloud service, vector database, or public career profile. Private career files are user-controlled local artifacts and should not be committed.

VitaeGraph structures supplied material. It does not verify claims automatically, guarantee retrieval quality, or guarantee better platform outcomes.