What Are the Five Dimensions Scored in the A-H Evaluation?

The A-H evaluation in career-ops scores five distinct dimensions—Role Summary, CV Match, Level Strategy, Compensation Research, and Personalization/Interview Prep—each contributing to a single global score from 1 to 5.

The A-H evaluation is a structured job-offer analysis system used by the open-source career-ops tool. It generates a comprehensive report organized into blocks A through H, then distills this assessment into a unified 1-5 score based on five specific scoring dimensions. These dimensions are defined in the project's shared documentation files and drive the quantitative rating that helps users quickly compare opportunities.

---

The Five Scoring Dimensions Explained

According to modes/_shared.md, "the evaluation scores five dimensions, integrated into one global score of 1-5"【grep result, line 77】. These dimensions are distinct from the seven report blocks (A-G); block H presents the final scoring, while block G serves as a legitimacy check that does not influence the global score.

1. Role Summary

This dimension captures a concise TL;DR of the job posting, extracting:

• Archetype and domain classification
• Function and seniority level
• Remote work status
• Team size
• Culture-screen flag (indicating potential red flags)

The Role Summary provides the foundational context against which all other dimensions are evaluated.

2. CV Match

CV Match measures alignment between the candidate's CV and job requirements:

• Maps each job description requirement to exact lines in the CV
• Performs a gaps analysis for missing qualifications
• Quantifies overlap to produce a match percentage

This dimension directly impacts the feasibility score of pursuing the opportunity.

3. Level Strategy

Level Strategy analyzes seniority positioning:

• Compares the role's demanded level versus the candidate's natural level for their detected archetype
• Generates a "sell-senior" negotiation plan for stretching upward
• Prepares an "if-down-leveled" contingency strategy

This ensures the candidate enters negotiations with clear positioning tactics.

4. Compensation Research

This dimension conducts bounded salary research:

• Examines advertised salary ranges against market data
• Assesses company-type compensation reliability (startup vs. enterprise vs. public-sector pay structures)
• Generates HR-verification questions for offer-stage negotiations

The research is deliberately bounded to avoid excessive data gathering on low-probability opportunities.

5. Personalization / Interview Prep

The final dimension creates tailored application and interview materials:

• STAR+R stories mapped to JD requirements
• CV and LinkedIn customization plan
• Interview-plan (block F) linking each requirement to concrete anecdotes

This operationalizes the evaluation into actionable next steps.

---

Where the Five Dimensions Are Defined

The README.md enumerates these dimensions explicitly as "Role summary, CV match, level strategy, comp research, personalization, interview prep (STAR+R)"【grep result, line 125】. The architecture separates report generation from scoring: blocks A-G present findings, while block H collapses the five dimensions into the final 1-5 rating.

File	Contribution to Scoring Model
modes/oferta.md	Defines full A-H workflow and dimension weighting
modes/_shared.md	Establishes five-dimension framework and global score calculation
README.md	User-facing documentation of dimension purposes

---

Practical Usage: Viewing Dimension Scores

CLI Execution

# Run full evaluation on a job posting URL

career-ops oferta --url https://example.com/job/12345

# The output displays the global score derived from five dimensions:

# **Score:** 4.2 / 5

# Extract structured dimension data for automation

career-ops oferta --url https://example.com/job/12345 --output json | jq '.global_score, .dimensions'

Programmatic Access

// Node.js API integration for custom workflows
const { evaluateOferta } = require('career-ops/lib/offer');

(async () => {
  const result = await evaluateOferta({ 
    url: 'https://example.com/job/12345' 
  });
  
  console.log('Global score:', result.globalScore);
  console.log('Dimension breakdown:', result.dimensions);
  // dimensions: { roleSummary, cvMatch, levelStrategy, compResearch, personalization }
})();

The JSON output exposes individual dimension scores alongside the composite, enabling custom filtering or visualization pipelines.

---

Summary

• The A-H evaluation produces a structured report (blocks A-H) with scoring concentrated in block H.
• Five dimensions feed the global 1-5 score: Role Summary, CV Match, Level Strategy, Compensation Research, and Personalization/Interview Prep.
• These dimensions are defined in modes/_shared.md and documented in the README.md.
• Block G (legitimacy check) operates outside the scoring model—it flags suspicious postings without affecting ratings.
• Both CLI and programmatic interfaces expose per-dimension and global scores for automation.

---

Frequently Asked Questions

What does the global 1-5 score represent?

The global score is a weighted composite of five dimension ratings, where 1 indicates poor fit and 5 indicates exceptional alignment. It enables rapid comparison across multiple opportunities without reviewing full reports.

Why are there seven report blocks but only five scoring dimensions?

Blocks A-F present qualitative findings that inform the dimensions, while block G performs a binary legitimacy check. Only the five dimensions are mathematically integrated into the 1-5 score displayed in block H【grep result, line 77】.

Can I customize dimension weights in career-ops?

The current implementation in modes/oferta.md uses fixed weighting determined by the archetype detection. Custom weighting would require modifying the evaluation rules in the mode definition files.

How does the STAR+R methodology fit into the five dimensions?

STAR+R (Situation, Task, Action, Result, Reflection) stories are the deliverable of the Personalization dimension, providing concrete interview responses tied to each job requirement.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}