task-verifier

task-verifier is an agent for Claude Code from mussolene/1c_hbk_bsl. It costs 18 tokens per session (222 once invoked), scanned A, original, MIT.

A verification agent that independently checks the current repository and reruns relevant commands. It treats PASS as proven, FAIL as contradicted or broken, and UNKNOWN as not locally verifiable.

In plain words
What is it for?
Use it after implementation or repair to check acceptance criteria, reproduce remaining problems, and describe the smallest safe fix without modifying production code.
Why use it?
It provides a fresh assessment based on the files and results that exist now, rather than relying on earlier claims or another agent’s report.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: mentions AGENTS.md.

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/mussolene/1c_hbk_bsl/task-verifier
Clone the repo
git clone --depth 1 https://github.com/mussolene/1c_hbk_bsl

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for task-verifier

README.md
[![agentmods](https://agentmods.dev/badge/agents/mussolene/1c_hbk_bsl/task-verifier.svg)](https://agentmods.dev/agents/mussolene/1c_hbk_bsl/task-verifier)
Your own site
<a href="https://agentmods.dev/agents/mussolene/1c_hbk_bsl/task-verifier"><img src="https://agentmods.dev/badge/agents/mussolene/1c_hbk_bsl/task-verifier.svg" alt="Measured on agentmods" height="20"></a>
Per session 18 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 222 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 5.1 $0.00018 $0.00222
Opus 5 $0.00009 $0.00111
Sonnet 5 $0.00004 $0.00044
Haiku 4.5 $0.00002 $0.00022

Measured 6d ago against content hash ea7a71b2b2d6, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

task-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 6d 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.

.claude/agents/task-verifier.md · 25 lines

What it actually says

You are the task-verifier.

Use AGENTS.md and docs/oacs-development.md. OACS/ACS is the durable workflow surface; do not create or update repo-task-proof-loop files. Use floating context: accept compact OACS capsules as orientation, then rebuild proof from current repository state and focused command results.

Behavior:

  • You are not the implementer.
  • Judge current repository state and current command results, not prior chat claims.
  • Rerun relevant checks directly.
  • Treat sibling reports as hints, not proof.
  • PASS a criterion only if it is proven now.
  • Use FAIL when contradicted, broken, or incomplete.
  • Use UNKNOWN when it cannot be verified locally.
  • Do not modify production code.
  • Record verification outputs as ACS evidence when ACS is available.
  • Report non-PASS criteria with minimal reproduction steps and the smallest safe fix.
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. 6d ago First seen · 25 lines · 18 tokens per session scan A ea7a71b2b2d6

Subscribe to this mod's changes

task-verifier is an agent published in the GitHub repository mussolene/1c_hbk_bsl (9 stars, last pushed 11d ago), licensed MIT. It adds 18 tokens to every session and 222 once invoked, about $0.0001 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

bsl-verifier

Дешёвый верификатор BSL-кода после написания или правки .bsl. Прогоняет механический чеклист: сигнатуры методов платформы, существование и экспортность общих модулей, состав объектов метаданных, диагностики статического анализатора, именование #std454, допустимые символы. Возвращает вердикт и список нарушений с…

Romandredan/1c-quality-gate · 402 tokens

xml-runner

Дешёвый исполнитель проверок XML-выгрузки 1С. Прогоняет сверку «диск ↔ состав» и валидаторы структуры по изменённым файлам метаданных, разбирает их многословный вывод и возвращает вердикт с первыми ошибками по каждому файлу. Отличает недоступность инструмента от дефекта в проверяемом файле. НЕ правит XML и НЕ решает…

Romandredan/1c-quality-gate · 411 tokens

1c-code-explorer

Этот агент следует использовать, когда нужно глубоко проанализировать существующий код 1C: трассировать пути выполнения, понять архитектуру, найти паттерны и зависимости. Используй проактивно для глубокого анализа кодовой базы перед доработками.

AndreevED/1c-ai-feature-dev-workflow · 61 tokens

bsl-scout

Дешёвый разведчик по коду 1С для архитектурного контура. Собирает факты через индекс кода: кто вызывает метод, какие экспорты у общего модуля и как они кластеризуются по вызывающим, есть ли у экспорта триггеры в XML (подписки, команды, регламентные задания), состав объекта метаданных. Отвечает на вопросы «где / кто /…

Romandredan/1c-quality-gate · 451 tokens

1c-code-test-generator

Этот агент следует использовать, когда нужно создать тесты для кода 1C: юнит-тесты YaXUnit или сценарии Vanessa Automation (BDD). Используй проактивно после написания или изменения кода для покрытия тестами.

Diversus23/1c-feature-dev · 56 tokens

1c-code-explorer

Этот агент следует использовать, когда нужно глубоко проанализировать существующий код 1C: трассировать пути выполнения, понять архитектуру, найти паттерны и зависимости. Используй проактивно для глубокого анализа кодовой базы перед доработками.

Diversus23/1c-feature-dev · 60 tokens