What Are the Seven Scoring Blocks of the A-G Evaluation Framework in CareerOps?

The CareerOps A-G Evaluation Framework evaluates job opportunities through seven distinct scoring blocks—Role Fit (A), Compensation (B), Growth & Learning (C), Culture & Values (D), Team & Leadership (E), Impact & Responsibility (F), and Posting Legitimacy (G)—that aggregate into a global score to drive application decisions.

CareerOps is an open-source CLI tool designed to standardize how software professionals analyze job offers using structured data extraction. At its analytical core sits the A-G Evaluation Framework, a seven-block scoring methodology defined in the project's architecture documentation that converts qualitative job descriptions into quantitative decision metrics.

The Seven Scoring Blocks Explained

The framework is implemented in modes/oferta.md, which defines the evaluation prompts and scoring logic for each dimension. Every block receives a normalized score out of 5.0, contributing to a composite global rating.

Block A – Role Fit

Block A measures how precisely the role's responsibilities, required technical skills, and daily tasks align with the candidate's existing experience and stated career objectives. This block prevents applications to positions where the actual work diverges significantly from the job description or the candidate's expertise.

Block B – Compensation

Block B evaluates total cash compensation, equity packages, and bonus structures against current market benchmarks and the candidate's personal financial targets. This block flags offers that fall below market percentiles or fail to meet minimum compensation thresholds.

Block C – Growth & Learning

Block C assesses opportunities for professional development, including mentorship availability, formal training budgets, conference attendance policies, and defined career-path progression. High scores indicate environments invested in skill acquisition and leadership development.

Block D – Culture & Values

Block D examines company culture, organizational mission, diversity and inclusion practices, and work-life balance policies. This block validates alignment between corporate values and the candidate's personal ethical framework and lifestyle requirements.

Block E – Team & Leadership

Block E analyzes the quality of the hiring manager, existing team dynamics, leadership communication styles, and the team's historical delivery track record. Poor scores here indicate high-risk reporting structures or dysfunctional team environments.

Block F – Impact & Responsibility

Block F quantifies the scope of influence granted to the role, including strategic ownership, cross-functional impact, and the position's significance within the broader organization. This block distinguishes high-autonomy roles from narrow execution-focused positions.

Block G – Posting Legitimacy

Block G serves as a mandatory gate-check that verifies the job posting is currently live, authentic, and not a duplicate or expired listing. Unlike other blocks, this functions as a binary pass/fail prerequisite; if Block G fails, no other scoring blocks are evaluated.

How the A-G Framework Calculates Decision Scores

Each of the seven blocks generates an individual score between 0.0 and 5.0. According to ARCHITECTURE.md, these individual scores aggregate into a global A-G score that determines the recommended action. CareerOps applies a default threshold of ≥3.0/5 to trigger an "apply" recommendation, while scores below this threshold suggest passing on the opportunity or entering negotiation phases.

The evaluation pipeline processes blocks sequentially, with Block G executing first as a filtering mechanism. If the posting fails the legitimacy check, the CLI exits immediately without consuming API resources on the qualitative analysis of blocks A through F.

Running A-G Evaluations with the CareerOps CLI

Execute a full evaluation by passing a job description URL to the oferta mode. The tool generates a markdown report in the reports/ directory containing structured scores for all seven blocks.

# Execute a full A-G evaluation on a live job posting

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

# The command outputs reports/###-company-date.md containing

# individual block scores and the final global recommendation

The generated report follows a strict markdown schema where each block appears as a level-2 heading with a numeric score. You can programmatically extract these scores for further automation or CRM integration.

# Extract numeric scores from a generated report using Node.js

node -e "
import fs from 'fs';
const report = fs.readFileSync('reports/001-example-2024-08-22.md', 'utf8');
const scores = [...report.matchAll(/## Block ([A-G]) – .*?\\n\\n([\\d.]+)\\/5/g)]

  .reduce((obj, [, block, score]) => ({...obj, [block]: parseFloat(score)}), {});
console.log(scores);
"

Example output:

{
  "A": 4.5,
  "B": 3.8,
  "C": 4.2,
  "D": 4.0,
  "E": 3.9,
  "F": 4.3,
  "G": 5.0
}

Architecture and Source File References

The A-G Evaluation Framework is distributed across several key files in the santifer/career-ops repository:

• modes/oferta.md – Contains the full implementation of the seven-block evaluation, including specific prompts for each scoring dimension and the mandatory Block G posting legitimacy verification.
• ARCHITECTURE.md – Defines the high-level pipeline architecture where the "7 blocks (A-G)" integrate into the evaluation stage, specifying how individual block scores flow into the global decision algorithm.
• README.md – Provides user-level documentation for invoking the oferta mode and interpreting the resulting markdown reports.
• batch/batch-prompt.md – Describes the batch processing workflow that expects a completed A-G evaluation report for each job processed in bulk operations.

Summary

• The A-G Evaluation Framework consists of seven blocks: Role Fit (A), Compensation (B), Growth & Learning (C), Culture & Values (D), Team & Leadership (E), Impact & Responsibility (F), and Posting Legitimacy (G).
• Block G acts as a mandatory prerequisite gate that validates posting authenticity before qualitative scoring begins.
• Each block scores out of 5.0, feeding a global A-G score that uses a default 3.0/5 threshold for application recommendations.
• Evaluation logic resides primarily in modes/oferta.md, with architectural specifications in ARCHITECTURE.md.
• Reports generate as markdown files in the reports/ directory, enabling programmatic parsing and third-party pipeline integration.

Frequently Asked Questions

What makes Block G different from the other scoring blocks?

Block G (Posting Legitimacy) functions as a binary gate-check rather than a qualitative score. As implemented in modes/oferta.md, this block verifies the job URL is active and authentic before any API calls evaluate blocks A through F. If Block G detects an expired or duplicate listing, the framework terminates immediately with a zero score, conserving computational resources and preventing analysis of invalid opportunities.

How does CareerOps calculate the final global score from the seven blocks?

The framework aggregates individual block scores (each ranging 0.0–5.0) into a composite global A-G score. According to ARCHITECTURE.md, this aggregation follows a weighted or normalized algorithm that produces a single recommendation threshold. By default, a global score ≥3.0/5 triggers an "apply" status, while lower scores suggest rejection or negotiation, though individual block weaknesses may override the aggregate in specific workflow configurations.

Where does CareerOps store the evaluation reports after running the A-G framework?

Upon executing career-ops oferta <url>, the CLI writes structured markdown reports to the reports/ directory using the filename pattern ###-company-date.md (for example, 001-example-2024-08-22.md). These files contain individual block scores, qualitative justifications for each rating, and the final global recommendation, formatted for both human review and programmatic parsing.

Can the scoring weights be customized for each block in the A-G Evaluation Framework?

While the default implementation in modes/oferta.md treats all seven blocks as equally weighted inputs to the global score, the open-source architecture allows modification of the evaluation prompts and scoring algorithms. Advanced users can fork the repository and adjust the weighting logic in the evaluation mode file or post-process the markdown reports with custom scripts to apply non-standard aggregation formulas tailored to specific career priorities.

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