agentic-finance-review

Name: agentic-finance-review
Rating: 58 (1 reviews)
Author: disler

disler/agentic-finance-review

An implementation guide and framework for building specialized, self-validating agents using Claude Code hooks to automate financial data reviews.

99 45No licenseUpdated 2026-01-18

Fair

View on GitHub Open source file

Overview

This repository demonstrates how to implement specialized self-validating agents using Claude Code's hook system. By pairing hyper-focused subagents and slash commands with deterministic Python validation scripts (running on PreToolUse, PostToolUse, and Stop events), the framework ensures reliable automation for tasks like normalizing bank CSVs, merging datasets, generating financial visualizations, and building interactive HTML dashboards.

Capabilities

▸Custom slash commands for Claude Code
▸Specialized subagents with context isolation
▸Deterministic post-tool-use and stop validation hooks
▸Multi-agent financial data processing pipeline
▸Automated chart generation and HTML dashboard creation

Best for

Automating multi-account financial statement ingestion and normalization, Running parallel CSV edits with automated data integrity and balance validation, Generating year-to-date financial insight reports and interactive dashboards via AI agents

Works with

Claude Code

Evoa Score breakdown= Σ (score × weight)

Task usefulness20%88 → +17.6

実タスクにどれだけ役立つか（機能の豊富さ・用途の明確さ）。 — AIによるcapabilities/use-cases解析

Code quality15%90 → +13.5

実装・指示の品質。 — AIによるSKILL.md/README解析

Maintenance15%50 → +7.5

リポジトリがどれだけ活発に保守されているか。 — GitHub 最終push日時の新しさ

Documentation12%26 → +3.1

ドキュメントの充実度・分かりやすさ。 — README/独自要約の情報量

Security15%80 → +12.0

危険・不審な挙動が無いか。 — AIによるセキュリティレビュー

Originality10%85 → +8.5

ありふれたラッパーではない独自性。 — AIによる独自性判定

Popularity8%57 → +4.6

コミュニティの採用度。 — GitHub Stars/Forks（対数スケール）

Compatibility5%25 → +1.3

対応AIエージェントの広さ。 — AIによる対応エージェント判定

Weighted total68.0 / 100

ライセンス不明/制限あり（Red）のSkillは総合スコアに0.85倍の補正を適用します。ランキングはこのScoreのみで決まり、広告で変わりません。算出方法の詳細 →

Security considerations

Processes financial datasets. Users must ensure real bank statements are kept out of public repositories by configuring private forks, adding data directories to .gitignore, and avoiding the commit of sensitive configuration files.