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/dimkurilo/opencode-skills/project-bootstrapnpx skills add dimkurilo/opencode-skills --skill project-bootstrapgit clone --depth 1 https://github.com/dimkurilo/opencode-skillsWhat 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 | $0.00171 | $0.05720 |
| Opus 5 | $0.00086 | $0.02860 |
| Sonnet 5 | $0.00034 | $0.01144 |
| Haiku 4.5 | $0.00017 | $0.00572 |
Grade A, and why
project-bootstrap 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.
How it starts
The opening of the file, as written. The whole thing — 415 lines — stays where its author put it; the contents beside it link to each section on GitHub.
project-bootstrap v2.2.0
Создаёт агентскую инфраструктуру с адаптацией под тип проекта, модель и сложность. Variant E + GRACE-якоря — правила неизбежны: в преамбуле (primacy) и closing anchors (recency).
Быстрый старт: bash scripts/classify_project.sh → классификация → выбор шаблона → генерация → contradiction check → двойной аудит.
TL;DR (minimum path)
project-bootstrap = ДОМ для агента. Создаёт AGENTS.md (Variant E: правила в преамбуле + closing anchors), SESSION_HANDOFF.md, .gitignore, .agents/memory/MEMORY.md, rules. Адаптация под тип (ops/code/agent/content) и модель (DeepSeek/GLM/Qwen/universal).
Это точка входа. После bootstrap: план спринта → wave-spec; нужно ≥2 моделей → multi-model-orchestration.
Не использовать для мелких правок/аудита/уже настроенного проекта (там режим расширения).
Next step (вставить в сгенерированный AGENTS.md в Фазе 6) — роутер-таблица, чтобы владелец видел навигацию в файле проекта:
| Хочу… | Скилл |
|---|---|
| Новый проект / настроить среду | project-bootstrap |
| Спланировать и сделать спринт (волну) | wave-spec |
| Обсудить / отревьюить 2+ моделями | multi-model-orchestration |
Workflow
Фаза 0: Классификация проекта
- Запусти классификатор:
bash scripts/classify_project.sh <project_dir> - Прочитай вывод — JSON с полями:
type,complexity,variant,model_profile. - Проверь существующий AGENTS.md:
ls AGENTS.md 2>/dev/null— если найден → режим расширения (не перезаписывать, дополнять).
Типы проектов и приоритет гибридов (ops > code > agent > content):
| Тип | Сигналы | Шаблон |
|---|---|---|
ops / ops-code |
Docker, SSH, .env, nginx, продакшен-риски | Variant E полный (преамбула + чеклист + IF-THEN) |
code / code-content |
Исходники (py/ts/js), тесты, CI/CD | Variant E + GRACE-якоря |
agent |
Скиллы, агенты, промпты, модели | Variant E + модельно-специфичные CLOSING ANCHORS |
content |
Статьи, briefs, Excel, интеграции | Облегчённый: преамбула без тех. правил |
undetermined |
<3 файлов, нет AGENTS.md | Базовый шаблон с предложением доработать |
What ships with it
27 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.
- assets/hero-dark.svg 1.5 KB
- assets/hero-light.svg 1.4 KB
- assets/templates/agent-persona.md.tmpl 770 B
- assets/templates/AGENTS.md.tmpl 4.4 KB
- assets/templates/api-config.example.tmpl 380 B
- assets/templates/command.md.tmpl 354 B
- assets/templates/general-rule.md.tmpl 7.2 KB
- assets/templates/MEMORY.md.tmpl 2.9 KB
- assets/templates/nda-anonymization.md.tmpl 2.6 KB
- assets/templates/opencode-agent.md.tmpl 472 B
- assets/templates/plan.md.tmpl 2.3 KB
- assets/templates/rule.md.tmpl 722 B
- assets/templates/script.py.tmpl 767 B
- assets/templates/script.sh.tmpl 227 B
- assets/templates/SESSION_HANDOFF.md.tmpl 1.9 KB
- assets/templates/SKILL.md.tmpl 499 B
- assets/templates/YYYY-MM-DD.md.tmpl 422 B
- README.md 4.4 KB
- README.ru.md 5.9 KB
- references/grace-anchors.md 5.1 KB
- references/model-profiles.md 25 KB
- references/operational-rules.md 9.8 KB
- references/playbook.md 8.1 KB
- references/variant-e-structure.md 7.8 KB
- references/workflow-patterns.md 4.2 KB
- scripts/classify_project.sh 5.8 KB runs code
- scripts/verify-handoff-gate.sh 3.8 KB runs code
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.
- 2d ago First seen · 415 lines · 171 tokens per session scan A 9fe1abb9ea4d
project-bootstrap is a skill published in the GitHub repository dimkurilo/opencode-skills (6 stars, last pushed 19d ago), licensed MIT. It adds 171 tokens to every session and 5,720 once invoked, about $0.0009 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.
Other skills, from other repositories
agent-messaging
Send and receive cryptographically signed messages between AI agents using the Agent Messaging Protocol (AMP). Use when the user asks to "send a message to an agent", "check agent inbox", "message another agent", "reply to a message", "notify an agent", or any inter-agent communication task.
kagent-dev
Development guide for kagent's v1alpha3 Harness and AgentTemplate CRDs, AgentInstance gRPC control plane, upstream A2A integration, Substrate runtime provisioning, tests, generation, and PR workflow. Use for any implementation, debugging, review, or CI task in the kagent repository.
workflows
Execute a workflow script that orchestrates multiple subagents deterministically. Use when the user explicitly requests a workflow, multi-agent orchestration, parallel or sequential agent pipelines, structured agent outputs, or a durable background workflow run.
agent-tuning
Use when changing agent model or effort configuration, adapter mappings, or eval candidate profiles.
ax-python-agent
Use when writing Python code with axllm for agents, child delegation, tools, MCP, citations, persistent playbook learning, stage instructions, runtime state, final typed responses, and direct-respond executor skipping.
📝 任务完成后归档
重要提醒: 每次完成复杂调试或开发任务后，主动执行此流程！ 将学到的经验归档为 skill，供以后参考。不要等用户提醒。.