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/vynazevedo/first-plan/protocolnpx skills add vynazevedo/first-plan --skill protocolgit clone --depth 1 https://github.com/vynazevedo/first-planWhat 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 | $0.00081 | $0.01687 |
| Opus 5 | $0.00041 | $0.00843 |
| Sonnet 5 | $0.00016 | $0.00337 |
| Haiku 4.5 | $0.00008 | $0.00169 |
Grade A, and why
first-plan-protocol 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 — 167 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Protocol - first-plan
Skill base obrigatória do plugin. Toda execução de comando /fp:* carrega esta skill.
Filosofia
Antes de escrever uma linha de código, entenda o projeto. O plugin existe pra forçar essa disciplina. Não improvisar, não impor padrões externos, não duplicar trabalho.
Protocolo PLAN-FIRST (obrigatório)
Discovery -> Plan -> Approval -> Execution -> Report
Fases bloqueantes. Cada uma só inicia quando a anterior estiver concluída e validada.
Fase 1 - Discovery
Disparada por /fp:init. Resultado: .first-plan/ populado.
Mapear:
- Topologia (stacks, arquitetura, boundaries, deployment)
- Convenções (naming, errors, testing, logging, di, security)
- Reuse Index invertido (componentes, utils, types, hooks)
- Domínio (glossário, entidades, flows)
- Riscos (fragile, untested, magic, debt)
- Rationale (do, dont, why)
- Spec-Code Reconciliation (matriz feature × status × evidência)
- Git intelligence (activity, ownership, in-flight)
Cada finding tem confidence: 0.0-1.0. Findings com confidence < 0.7 viram entradas em .first-plan/08-meta/questions.md.
Fase 2 - Plan
Disparada por /fp:plan <feature>. Resultado: .first-plan/07-state/plans/<slug>.md.
Antes de escrever o plano:
- Consultar
.first-plan/09-features/- a feature ja existe? - Consultar
.first-plan/07-state/in-flight.md- alguém ja esta fazendo? - Consultar
.first-plan/03-reuse/INDEX.md- o que reusar?
O plano deve seguir o template em ${CLAUDE_PLUGIN_ROOT}/meta-templates/plan.md. Pausa pedindo aprovação humana ao final.
Fase 3 - Approval (gate humano)
Estado: awaiting_approval em .first-plan/07-state/STATE.md.
Aguardar:
/fp:execute→ prossegue- Mensagem com mudança → reescreve plano e volta a aguardar
Não executa nada antes da aprovação. Não há atalho.
Fase 4 - Execution
Disparada por /fp:execute. Resultado: código modificado conforme plano + atualização do STATE.
Durante:
- Seguir o plano à risca, na ordem definida
- Commits pequenos e atômicos no padrão do projeto
- Se descobrir algo que invalide o plano (premissa errada, contrato diferente, erro de discovery): PARAR e reportar, não improvisar
- Atualizar STATE.md a cada passo
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 · 167 lines · 81 tokens per session scan A 0fd034156921
first-plan-protocol is a skill published in the GitHub repository vynazevedo/first-plan (23 stars, last pushed 8d ago), licensed MIT. It adds 81 tokens to every session and 1,687 once invoked, about $0.0004 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 2026-08-30.
Other skills, from other repositories
RecallMax
Enhances AI memory capabilities with long-context support.
agent-memory
Persistent memory system for AI coding agents. Use when you need to save facts about the user, store interaction preferences, log session events, or search previously saved memories. Triggers on remembering, saving context, recalling past decisions, or when the user says "remember this". Also use proactively when the…
snip
You are an expert at writing declarative YAML filters for snip, a CLI proxy that reduces LLM token consumption by filtering shell output.
release-notes
Draft user-facing release notes from a git log range. Trigger when the user asks for "release notes", "changelog entry", or "what shipped this week" with a git ref range. Output is markdown sections grouped by Features / Fixes / Breaking with PR links.
arch-check
架构与实现审查 —— 基于「概念建模 → 职责划分 → 机制/策略分离 → 因果与不变量 → 属性建模 → 模块化 → SOLID → GRASP → YAGNI」的全维度审查，带置信度门控与假阳性抑制（对抗"过度工程建议"这类 AI slop）。触发于：要求 review/审查架构、检查目录结构/依赖关系/职责划分、重构前评估、技术债盘点、判断是否过度设计，或问「这个设计合理吗 / 该怎么拆 / 有没有循环依赖 / 这个改动架构上 OK 吗」。一律按最高强度审查。用法：arch-check [范围] [--fix|--plan]。范围可为目录/文件/PR；缺省审当前分支相对基线的全部变更。.
Agent Context File Writer
Writes a high-quality CLAUDE.md, .cursorrules, or .windsurfrules file that gives a coding agent the right project context, conventions, and constraints to work effectively.