# Agent Parser

> End-to-end resume parsing (detect format → extract fields). Uses a combination of format detection, text extraction, and LLM parsing to normalize resume data.

- Skill: `diegosouzapw/agent-parser` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add diegosouzapw/agent-parser`
- Raw SKILL.md: https://api.skillmd.com/api/skills/diegosouzapw/agent-parser/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: diegosouzapw (https://skillmd.com/u/diegosouzapw)
- Updated: 2026-09-08
- Page: https://skillmd.com/skills/diegosouzapw/agent-parser

---


# Parser Agent

## Overview

The Parser Agent handles the ingestion of resume files. It follows a multi-step workflow:
1.  Detects file format (`detect-resume-format`)
2.  Extracts raw text (`extract-text-pdf`, `extract-text-docx`)
3.  Sanitizes text (`sanitize-text`)
4.  Parses identifying fields (`llm-parse-resume`)

## Workflow Definition

1.  **Input**: Resume file path.
2.  **Detection**: Call `detect-resume-format` script.
3.  **Extraction**:
    *   If PDF: Call `extract-text-pdf`
    *   If DOCX: Call `extract-text-docx`
    *   If Unknown/Txt: Read file directly
4.  **Sanitization**: Call `sanitize-text`.
5.  **Parsing**: Call `llm-parse-resume`.
6.  **Validation**: Call `schema-validate-resume`.
7.  **Output**: Validated JSON object.