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 skills/alekseiul/openclaw-superagent/agent-forgenpx skills add AlekseiUL/openclaw-superagent --skill agent-forgegit clone --depth 1 https://github.com/AlekseiUL/openclaw-superagentWrote 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/skills/alekseiul/openclaw-superagent/agent-forge)<a href="https://agentmods.dev/skills/alekseiul/openclaw-superagent/agent-forge"><img src="https://agentmods.dev/badge/skills/alekseiul/openclaw-superagent/agent-forge.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 | $0.00111 | $0.06181 |
| Opus 5 | $0.00056 | $0.03090 |
| Sonnet 5 | $0.00022 | $0.01236 |
| Haiku 4.5 | $0.00011 | $0.00618 |
Grade A, and why
agent-forge 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.
This is a copy
95% identical to agent-forge — 9 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 526 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AgentForge 🔧
Создание и улучшение скиллов и агентов OpenClaw. Лучшие практики + боевой опыт с десятками скиллов и агентов. Агенты: от базового (5 мин) до полноценного рабочего (с памятью, автоулучшением, командой).
Шаблоны файлов агента: references/agent-templates.md
Быстрый старт
Скилл за 5 минут:
mkdir skills/<имя>/→ создатьSKILL.mdс frontmatter (name, description) + алгоритм + 2 примера- Готово. Скилл подхватится автоматически (hot-reload)
Агент за 5 минут:
- Добавить в
agents.list[]в openclaw.json (id, name, model, workspace, tools.deny) mkdir -p ~/.openclaw/agents/<id>/agent+AGENTS.mdс ролью- Перезапустить gateway (через config.patch или
openclaw gateway restart)
Нужно по шагам? Читай дальше.
Режим A: СКИЛЛ
Шаг 1. Онбординг (2-4 вопроса, по одному)
- Что скилл должен делать? Конкретный пример использования
- Когда активировать? (триггерные фразы)
- Нужны ли скрипты, данные, внешние API?
- Публичный (для подписчиков) или внутренний?
Шаг 2. Определи тип скилла
Workflow (пошаговый процесс) - если задача = последовательность действий. → Пример: deep-research-pro (5 шагов: уточнение → поиск → синтез → отчёт)
Role (экспертная роль) - если задача = "отвечай как специалист X". → Пример: copywriter (пиши как владелец, вот стиль, вот примеры)
Data-driven (работа с данными) - если нужны конкретные факты/профили. → Пример: auto-mechanic (профиль машины в data/, логика диагностики в SKILL.md)
Гибрид - комбинация. Пример: family-doctor = Role + Data (роль врача + медпрофили в data/).
Шаг 3. Планирование структуры
В зависимости от типа (шаг 2):
- Workflow → обычно хватит одного
SKILL.md(вся логика помещается в core) - Role →
SKILL.md+references/(словарь стиля, примеры голоса, правила) - Data-driven →
SKILL.md+data/(профили, базы, справочники) - Гибрид →
SKILL.md+references/+data/(по необходимости)
Также могут понадобиться:
scripts/- Python/Bash для повторяемых операцийassets/- шаблоны, изображения
What ships with it
11 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- banner.jpg 123 KB
- examples/code-reviewer/SKILL.md 3.1 KB
- examples/content-planner/references/style-guide.md 1.7 KB
- examples/content-planner/SKILL.md 3.5 KB
- examples/meeting-prep/SKILL.md 3.3 KB
- examples/task-tracker/data/tasks.md 367 B
- examples/task-tracker/SKILL.md 2.5 KB
- examples/weather-bot/SKILL.md 1.6 KB
- LICENSE 1.1 KB
- README.md 5.5 KB
- references/agent-templates.md 13 KB
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.
- 6d ago First seen · 526 lines · 111 tokens per session scan A cb1da671062d
agent-forge is a skill published in the GitHub repository AlekseiUL/openclaw-superagent (40 stars, last pushed 3mo ago), licensed MIT. It adds 111 tokens to every session and 6,181 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. It is 95% identical to agent-forge, differing in 9 lines, and is treated as a copy.
Other skills, from other repositories
stage-assemble
How to execute a validated, currently authorized project/plan.json into one finished file. Start/resume with production.status; if planapprovalcurrent is false, pass the state to gate-control and stop instead of producing or opening child gates. Walk the signed EDL; do not re-plan. Host-neutral: VideoStudio-specific…
stage-plan
How to turn "here is my material + here's the video I want" into a single, inspectable plan that spans reference images, reference videos, deterministic editing, semantic editing, generation, and composition. The output is project/plan.json — a cross-modal Edit Decision List (EDL) — which the assembler walks.…
ppt-planner
Read only after ppt-router chooses CREATE or OUTLINE. Planning is the source of truth for construction, not optional prose around the deck.
ppt-review
Read this skill only after current officereview structural output and relevant rendered images exist. Do not use source code, a tool success flag, or the first-slide preview as a substitute for deck evidence.
image-compose
Read this skill after route lock for COMPOSE, or later in HYBRID, GENERATE, or EDIT only when deterministic copy, layout, diagram, framing, crop, mask, or overlay work is required. Do not preload it during routing. Keep the artifact in one stable directory containing image-manifest.json, index.html, and local assets.
image-router
Read this skill first and alone for every image creation or revision request. Return a locked route before reading any other ImageStudio skill; do not batch this read with craft, canvas, production, or review skills.