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 rules/dewtech-technologies/dare-method/skill-dag-runnergit clone --depth 1 https://github.com/dewtech-technologies/dare-methodWrote 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/rules/dewtech-technologies/dare-method/skill-dag-runner)<a href="https://agentmods.dev/rules/dewtech-technologies/dare-method/skill-dag-runner"><img src="https://agentmods.dev/badge/rules/dewtech-technologies/dare-method/skill-dag-runner.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 | $0.00000 | $0.02507 |
| Opus 5 | $0.00000 | $0.01254 |
| Sonnet 5 | $0.00000 | $0.00501 |
| Haiku 4.5 | $0.00000 | $0.00251 |
Grade A, and why
skill-dag-runner 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 today.
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 — 222 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill: DAG Task Runner
Quando usar
- O
BLUEPRINT.mdfoi aprovado e você precisa decompor em tasks executáveis - O usuário pediu para começar a execução do DAG
- Existe
DARE/dare-dag.yamle você precisa entender, executar ou modificar - Aparece o canvas
DARE/.canvas.mddurante uma execução
Quem executa o quê
Você (Cursor) é o executor. O CLI
dareé orquestrador.
- A IDE em que você roda já está autenticada na conta do usuário
- Você lê o
dare-dag.yamle as specs emDARE/EXECUTION/task-*.md - Você executa cada task — escreve código, roda testes, faz lint
- Depois de cada task, você chama o CLI para registrar o resultado:
dare execute --complete <task-id> --output "<resumo>"dare execute --fail <task-id> --reason "<mensagem>"
- O CLI atualiza o canvas e popula o
dare-graphautomaticamente
Não há API key. Não há custo extra de tokens. Você usa o plano da IDE.
O que é o DAG do DARE
DARE/dare-dag.yaml é o plano de execução: um grafo direcionado acíclico
em que cada nó é uma task atômica e as arestas são depends_on. O CLI ordena
topologicamente (Kahn's algorithm) e indica em que ordem você deve executar.
Tasks no mesmo rank podem rodar em paralelo (logicamente — você decide se
literalmente fan-out ou roda uma após a outra).
rank 0: task-001 task-002 ← podem rodar juntas (sem depends_on)
rank 1: task-003 (deps: 001, 002) ← só após rank 0
rank 2: task-004 (deps: 003)
Schema canônico do dare-dag.yaml
title: "<Nome do projeto> - Development Tasks"
version: "1.0.0"
limits:
parent_context_chars: 2000 # snippet de cada output de pai injetado no filho
task_output_chars: 4000 # cap do output capturado por task
timeout_seconds: 600 # apenas referência; quem aborta é você
# Mapeamento complexity → modelo (referencial — você usa o modelo da sua IDE)
models:
cursor: { HIGH: gpt-5.3-codex, MED: composer-2, LOW: auto-low }
claude: { HIGH: claude-sonnet-4-5, MED: claude-haiku-4, LOW: claude-haiku-4 }
antigravity: { HIGH: gemini-2.5-pro, MED: gemini-2.5-flash, LOW: gemini-2.5-flash }
tasks:
- id: task-001
title: "Setup project structure"
depends_on: []
complexity: LOW
spec_file: EXECUTION/task-001.md
subtask_prompt: |
<prompt completamente self-contained>
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.
- today First seen · 222 lines · 0 tokens per session scan A 582acb94d661
skill-dag-runner is a cursor rule published in the GitHub repository dewtech-technologies/dare-method (5 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,507 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 2026-09-03.
Other cursor rules, from other repositories
hedgehog
Hedgehog build discipline — how to work in this repo.
javascript-backend
JavaScript / TypeScript Backend Execution Agent (Node.js / Express / Fastify / NestJS).
java-backend
Java Backend Execution Agent (Spring Boot / Quarkus).
observability
Observability Standards — Logging, Tracing & Metrics.
api-design
REST API Design Standards Agent.
pr-standards
Git Commit & PR Standards for AI-Assisted Development.