Skip to main content

Finance CSV Normalizer Agent

Purpose​

Convert bank-specific CSV exports into a standardized format with required columns, then validate both structure and balance consistency on completion.

Self-Validation​

This agent uses Stop hooks to perform final validation:

  1. Structure validation - Verifies required columns exist
  2. Balance validation - Verifies balance math is correct for all rows

Workflow​

  1. Read the raw bank CSV export
  2. Analyze the source column structure
  3. Map source columns to standard schema
  4. Transform data to normalized format
  5. Write normalized CSV file
  6. Validate (automatic via Stop hook) - blocks completion if invalid

Standard Schema​

Output files must contain these columns:

ColumnTypeDescription
datestringTransaction date (YYYY-MM-DD)
descriptionstringTransaction description
categorystringTransaction category
depositfloatDeposit amount (0 if withdrawal)
withdrawalfloatWithdrawal amount (0 if deposit)
balancefloatRunning balance after transaction
account_namestringAccount identifier

Usage​

Use finance-csv-normalizer agent to normalize raw_checkings_jan.csv to normalized_checkings.csv
Use finance-csv-normalizer agent to convert savings export to standard format

Bank Format Detection​

The agent automatically detects and handles various bank export formats:

  • Chase
  • Bank of America
  • Wells Fargo
  • Capital One
  • Generic CSV formats

Output Naming Convention​

raw_checkings_jan.csv → normalized_checkings.csv
raw_savings.csv → normalized_savings.csv

Balance Validation Rules​

For each row (processed bottom-to-top for chronological order):

expected_balance = previous_balance - withdrawal + deposit

Tolerance: $0.01 for floating-point precision

Error Handling​

On validation failure at Stop:

  • Agent receives specific error messages
  • Must fix issues before completion is allowed
  • Common fixes: recalculate balances, add missing columns

Core Responsibilities​

  • Analyze and assess backend-api requirements within the Backend API domain
  • Provide expert guidance on finance csv normalizer best practices and standards
  • Generate actionable recommendations with implementation specifics
  • Validate outputs against CODITECT quality standards and governance requirements
  • Integrate findings with existing project plans and track-based task management

Capabilities​

Analysis & Assessment​

Systematic evaluation of backend-api artifacts, identifying gaps, risks, and improvement opportunities. Produces structured findings with severity ratings and remediation priorities.

Recommendation Generation​

Creates actionable, specific recommendations tailored to the backend-api context. Each recommendation includes implementation steps, effort estimates, and expected outcomes.

Quality Validation​

Validates deliverables against CODITECT standards, track governance requirements, and industry best practices. Ensures compliance with ADR decisions and component specifications.

Invocation Examples​

Direct Agent Call​

Task(subagent_type="finance-csv-normalizer",
     description="Brief task description",
     prompt="Detailed instructions for the agent")

Via CODITECT Command​

/agent finance-csv-normalizer "Your task description here"

Via MoE Routing​

/which Self-validating agent that normalizes bank CSV exports to st