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.
npx agentmods add agents/prikotov/task-agents-playbook/metricsgit clone --depth 1 https://github.com/prikotov/task-agents-playbookWrote 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.
[](https://agentmods.dev/agents/prikotov/task-agents-playbook/metrics)<a href="https://agentmods.dev/agents/prikotov/task-agents-playbook/metrics"><img src="https://agentmods.dev/badge/agents/prikotov/task-agents-playbook/metrics.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.01131 |
| Opus 5 | $0.00000 | $0.00566 |
| Sonnet 5 | $0.00000 | $0.00226 |
| Haiku 4.5 | $0.00000 | $0.00113 |
Grade A, and why
metrics 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 5d 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.
How it starts
The opening of the file, as written. The whole thing — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Метрики и статистика использования AI-агентов
В этом документе описывается процесс сбора статистики и анализа эффективности использования AI-агентов (Gemini, Claude, etc.) в проекте.
Цель
Отслеживание динамики продуктивности разработки с использованием ИИ. Основная задача — минимизировать количество итераций (правок) на одну задачу, повышая качество контекста и промптов.
Инструменты сбора статистики
1. Общая статистика по репозиторию
Скрипт bin/stat-by-week.sh собирает данные из git log по неделям.
- Метрики: Добавленные/удаленные строки, количество коммитов, количество измененных файлов.
- Фильтрация: Исключает автогенерируемые файлы (Phoenix assets, public, styles и т.д.).
- Команда:
bin/stat-by-week.sh
2. Статистика по PR агентов
Скрипт bin/agent-prs-by-week.sh использует GitHub CLI (gh) для анализа смерженных PR с определенной меткой (например, gemini-cli).
- Метрики: Количество PR, добавленные/удаленные строки, количество измененных файлов.
- Команда:
bin/agent-prs-by-week.sh <agent-label> # Пример: bin/agent-prs-by-week.sh gemini-cli
Таблица метрик
Данные агрегируются в Google Sheets.
Основные метрики:
- Lines Added/Deleted: Объем кода.
- PR Count: Количество завершенных задач.
- Commits per PR: (Вычисляемая) Среднее количество коммитов в одном PR.
Проблематика метрики "Commits per PR"
Текущая метрика "Commits per PR" показывает значения 1-3, что является "Ложноположительным" показателем качества, так как:
- Локальные итерации: Основная работа по исправлению замечаний происходит локально (в IDE) до пуша в репозиторий.
- Сквош коммитов: Агенты могут изменять существующие коммиты или пользователь делает squash перед мержем.
В результате метрика отражает чистоту истории, а не количество итераций "Промпт -> Правка".
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.
- 5d ago First seen · 70 lines · 0 tokens per session scan A ca4e4f136889
metrics is an agent published in the GitHub repository prikotov/task-agents-playbook (21 stars, last pushed 3mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,131 tokens. 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 2026-08-30.
Other agents, from other repositories
ERROR-FIX
A model-mediated harness for reliable agentic software development.
namba-designer
Use this role when art direction, palette/tone logic, composition, motion intent, or generic-looking UI surfaces need to be clarified before implementation or review. State effect: read-only. Do not edit repository files.
namba-backend-architect
Use this role when backend contracts, service boundaries, or persistence changes need to be clarified before implementation. State effect: read-only. Do not edit repository files.
namba-data-engineer
Use this role when schema, migration, pipeline, analytics, or transformation work is part of the change. State effect: workspace-write. Edit only files assigned by the main session.
agent-builder
Generates a new, complete Claude Code sub-agent configuration file from a user's description. Use this to create new agents. Use this Proactively when the user asks you to create a new sub agent, or uses the '/agents' command. When you prompt this agent, include the user's prompt VERBATIM. Remember, this agent has no…
amby-pm
Product Manager — AmbyKit role for specify/clarify/revise; use for that perspective.