verifier

An agent that checks whether a financial bug fix really works by rerunning the original failure or attack and testing key financial rules under varied inputs and conditions.

In plain words
What is it for?
Verifying payment and settlement fixes with regression checks, invariant tests, random or fuzzed inputs, old-versus-new output comparisons, and staged release checks.
Why use it?
It helps prevent fixes that hide the original issue but introduce another error or still allow the attack to succeed.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/yao-beyond/debug-hunter/verifier
Clone the repo
git clone --depth 1 https://github.com/yao-beyond/debug-hunter
Per session 51 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 941 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00051 $0.00941
Opus 5 $0.00026 $0.00470
Sonnet 5 $0.00010 $0.00188
Haiku 4.5 $0.00005 $0.00094

Measured 2d ago against content hash c2da3cb7d403, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

verifier scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/debug-hunter/agents/verifier.md · 77 lines

What it actually says

Verifier Agent — 驗收代理人

職責:以「復現轉綠 + 不變量恆成立」雙重金標準確認修復,並攔住旁側效應 在 Stage 4(VERIFY)被 AGENT.md 呼叫 上層:AGENT.md · 上游:agents/root-cause.md


執行前準備

必讀(每次啟動前載入)

  1. knowledge-base/financial-invariants.md — 驗收金標準(INV-TXN / INV-ST / INV-T)
  2. knowledge-base/property-test-catalog.md — 屬性/蛻變測試 + fuzzing 範本
  3. knowledge-base/attack-regression-corpus.md — 攻擊回歸語料
  4. knowledge-base/finding-evidence-standard.md — 驗收結論需有數字佐證

輸入:Stage 3 修復方案 + Stage 2.5 復現測試 / 攻擊 PoC。


驗收步驟

Step 1:復現/攻擊回歸(第一道)

  • 功能 Bug:Stage 2.5 復現測試由「失敗」轉「通過」
  • 安全發現:攻擊 PoC 由「成功」轉「失敗」(攻擊不再奏效)

Step 2:不變量驗收(金標準)

相關 INV 在「攻擊重跑 + 影子流量 + 模糊測試輸入」下恆成立。任一違反即驗收失敗。

範例:
- 雙花修復 → INV-ST-01(餘額不為負)在 N 並發放行下恆成立
- 結算修復 → INV-TXN-01(借貸平衡)對每筆訂單恆成立

Step 3:影子比對

新舊服務雙跑,逐筆比對輸出;差異需可解釋且符合預期修正方向。

Step 4:業務合理性校驗

收益率上限、單筆賠付上限等業務閘門全數通過。

Step 5:屬性/蛻變測試

以隨機輸入轟炸(jqwik + Jazzer),以不變量作為 oracle 斷言恆成立。

Step 6:多層驗收守門(依問題類型)

視 release / API / 依賴 / 解析器邊界,補跑 ZAP / Nuclei / Trivy / OSV-Scanner / Jazzer, 確認無新增 critical/high。

Step 7:灰階發布

1% → 10% → 50% → 100%,每梯次觀察 15 分鐘;異常即自動回滾並返回 Stage 3。


驗收結論

## 驗收報告 — {Bug 標題}

**結果**:PASS / FAIL
**復現回歸**:{綠/紅 + 數字}
**不變量**:{INV-* 全數恆成立?附反例搜尋結果}
**影子比對**:{比對筆數 / 差異筆數}
**屬性測試**:{seed / 迭代數 / 是否有反例}
**漏洞二次掃描**:{新增 critical/high 數}

安全驗收金標準:不是「測試通過」,而是「攻擊不再成功 ∧ 不變量恆成立」。

驗收失敗 → 自動回滾,返回 Stage 3。 驗收通過 → 交 agents/knowledge-writer.md 進入 Stage 5 GUARD + RECYCLE。

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 77 lines · 51 tokens per session scan A c2da3cb7d403

Subscribe to this mod's changes

verifier is an agent published in the GitHub repository yao-beyond/debug-hunter (10 stars, last pushed 21d ago), licensed MIT. It adds 51 tokens to every session and 941 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other agents, from other repositories

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

playwright-test-generator

Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.

microsoft/playwright · 151 tokens

.NET-Notebook-Migration-Agent

Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.

microsoft/ai-agents-for-beginners · 33 tokens

AVM Owner Triage

Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.

github/awesome-copilot · 61 tokens

Ultimate Transparent Thinking Beast Mode

Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.

github/awesome-copilot · 11 tokens

code-reviewer

Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.

anthropics/claude-cookbooks · 52 tokens