cc-viewer 1.6.232 → 1.6.233

package/concepts/pt-BR/CustomUltraplanExpert.md

@@ -0,0 +1,165 @@
+# Especialista UltraPlan personalizado — Guia de criação
+
+## O que fazem os dois campos de entrada
+
+- **Nome do especialista**: o rótulo exibido no botão de função na linha de variantes do UltraPlan (máx. 30 caracteres). É apenas um nome de exibição e **nunca** é enviado ao Claude Code.
+- **Corpo do prompt**: sua instrução de função. No momento do envio, o cc-viewer **automaticamente** o envolve em tags `<system-reminder>...</system-reminder>` com um cabeçalho de escopo `[SCOPED INSTRUCTION]`. Portanto, **escreva apenas o corpo** — não adicione tags `<system-reminder>` por conta própria.
+
+---
+
+## Como é o modelo do especialista?
+
+Cada especialista integrado (Code Expert / Research Expert) é essencialmente um bloco `<system-reminder>` injetado no contexto do Claude Code. Seu especialista personalizado passa exatamente pelo mesmo pipeline. Aqui está o modelo do **Research Expert** detalhado:
+
+```xml
+<system-reminder>
+[SCOPED INSTRUCTION] The following instructions are intended for the next 1–3
+interactions. Once the task is complete, these instructions should be gradually
+deprioritized and no longer influence subsequent interactions.
+
+Pre-requisite: Use `AskUserQuestion` to clarify the research scope, target
+audience, and deliverable format whenever the user's intent is ambiguous. Skip
+only if the intent is unambiguous.
+
+Leverage a multi-agent exploration mechanism to formulate an exceptionally
+detailed implementation plan.
+
+Instructions:
+1. Utilize the Agent tool to spawn parallel agents that simultaneously explore
+   various facets of the requirements:
+   - If necessary, deploy a preliminary investigator to conduct an initial
+     survey of industry-specific solutions using `webSearch`;
+   - If necessary, deploy a specialized investigator to research authoritative
+     sources—such as academic papers, news articles, and research reports—
+     using `webSearch`;
+   - Assign an agent to synthesize the target solution, while simultaneously
+     verifying the rigor and credibility of the gathered papers, news, and
+     research reports;
+   - If necessary, assign an agent to analyze competitor data to provide
+     supplementary analytical perspectives;
+   - If necessary, assign an agent to handle the implementation of a product
+     demo (generating outputs such as HTML, Markdown, etc.);
+   - If the task is sufficiently complex, you may assign additional teammates
+     to the roles defined above, or introduce other specialized roles; you are
+     permitted to schedule up to 5 teammates concurrently.
+
+2. Synthesize the findings from the aforementioned agents into a comprehensive,
+   step-by-step implementation plan.
+
+3. Utilize the Agent tool to spawn a set of parallel review agents; these
+   agents shall scrutinize the plan from multiple roles and perspectives to
+   identify any omitted steps and to propose reasonable additions or
+   optimizations.
+
+4. Consolidate the feedback received from the review agents, then invoke
+   `ExitPlanMode` to submit your final plan.
+
+5. Upon receiving the result from `ExitPlanMode`:
+   - If Approved: Proceed to execute the plan within this current session.
+   - If Rejected: Revise the plan based on the provided feedback, and then
+     invoke `ExitPlanMode` once again.
+   - If an Error Occurs: Do *not* follow the suggestions; prompt the user for
+     further instructions.
+
+Your final plan must include the following elements:
+- A clear summary of the proposed implementation strategy;
+- An ordered list of files to be created or modified, specifying the exact
+  changes required for each;
+- A step-by-step sequence for executing the implementation;
+- Identification of potential risks and corresponding mitigation strategies;
+- Creative ideation and suggestions for advanced enhancements;
+- If a product demo was generated, place the corresponding demo output in an
+  appropriate location and notify the user.
+</system-reminder>
+```
+
+---
+
+## Análise seção por seção
+
+### 1. Cabeçalho de escopo `[SCOPED INSTRUCTION]` (wrapper — gerado automaticamente)
+> The following instructions are intended for the next 1–3 interactions...
+
+Isso diz ao Claude Code: **estas instruções estão ativas apenas para as próximas 1–3 rodadas**, depois se dissipam. Impede que a "persona do especialista" vaze para conversas não relacionadas posteriormente.
+
+**Esta linha é gerada automaticamente pelo cc-viewer. Você não precisa escrevê-la.**
+
+### 2. Definição da tarefa inicial (**isto é o que você deve reescrever**)
+> Leverage a multi-agent exploration mechanism to formulate an exceptionally detailed implementation plan.
+
+Este é o "sujeito-verbo-objeto" de todo o modelo: **diz ao Claude Code a postura e o objetivo**. O padrão "exploração multiagente + plano de implementação" se encaixa bem em tarefas de **engenharia de software / planejamento**, mas parece estranho para muitos outros domínios (revisão de conteúdo, análise de dados, copywriting, pesquisa de mercado, auditoria de conformidade…).
+
+**Recomendamos fortemente reescrever esta linha para o seu objetivo**, por exemplo:
+
+- **Revisor de conteúdo**: "Você é um revisor de conteúdo sênior especializado em {domínio}. Seu objetivo é identificar imprecisões factuais, inconsistências de tom e fragilidades estruturais no material fornecido."
+- **Analista competitivo**: "Conduza uma análise competitiva rigorosa para {categoria de produto}. Produza uma matriz de comparação, insights de posicionamento e recomendações estratégicas."
+- **Copywriter**: "Gere várias variantes criativas de texto para {cenário}, cada uma com posicionamento, tom e estratégia de chamada para ação distintos."
+
+### 3. Etapas do fluxo de trabalho (1–5 itens — **reduza ou estenda com base na complexidade**)
+
+O Research Expert tem 5 etapas: **explorar → sintetizar → revisar → enviar plano → executar**. Isso impõe "multiagente paralelo + revisão cruzada + aprovação do plano" — três camadas de rigor, apropriadas para tarefas de alto risco/amplo escopo, mas **exageradas para as leves**.
+
+- **Tarefa simples** (consulta única / pequena correção): elimine o despacho multiagente e a revisão; apenas "produza a resposta" em uma única etapa.
+- **Tarefa moderada**: mantenha "explorar → sintetizar → revisar"; elimine o passo a passo do ExitPlanMode; entregue o resultado diretamente.
+- **Tarefa complexa e cara** (grande refatoração, comparação de várias opções, pesquisa interdisciplinar): mantenha todas as 5 etapas, possivelmente adicione uma etapa de "modelo de risco" ou "matriz de comparação de opções".
+
+### 4. Subfunções na Etapa 1 (**adapte ao seu domínio**)
+
+O Research Expert lista 6 funções potenciais (explorador da indústria, pesquisador acadêmico, sintetizador + verificador de fatos, analista da concorrência, produtor de demo, slot de extensibilidade). **Reescreva esta lista para o seu cenário**:
+
+- **Redação**: "coletor de fontes + analista de estilo + verificador de fatos"
+- **Análise de dados**: "agente de limpeza de dados + agente de modelagem estatística + agente de visualização"
+- **Auditoria de código**: "agente de análise estática + auditor de cadeia de dependências + modelador de ameaças"
+
+### 5. Lista de verificação final de entregáveis (**alinhe com sua necessidade real**)
+
+> Your final plan must include the following elements: ...
+
+O modelo original lista 6 elementos de um "plano de implementação". Seu entregável pode ser algo completamente diferente:
+
+- Um **relatório de pesquisa** → "Resumo executivo / Metodologia / Principais conclusões / Limitações / Recomendações de ação"
+- Um **relatório de revisão** → "Lista de problemas / Classificação de gravidade / Sugestões de correção / Exemplos antes e depois"
+- Uma **matriz de comparação** → "Definições de dimensões / Rubrica de pontuação / Conclusões / Justificativa da recomendação"
+
+---
+
+## Dicas de criação (TL;DR)
+
+1. **Mantenha o wrapper**: a linha `<system-reminder>` + `[SCOPED INSTRUCTION]` é adicionada pelo cc-viewer — não repita.
+2. **Reescreva a frase inicial**: declare a função, o objetivo e o formato de saída em uma única linha.
+3. **Flexibilize o fluxo de trabalho**: 1–2 etapas para tarefas leves, o ciclo completo de 5 etapas apenas para as complexas.
+4. **Reescreva as subfunções da Etapa 1**: os padrões (artigos acadêmicos / concorrentes / demo) provavelmente não são o que você quer.
+5. **A "lista de verificação de entregáveis" final é seu padrão de qualidade**: especifique a estrutura de saída — o Claude Code seguirá rigorosamente.
+
+---
+
+## Um exemplo refatorado: Analista competitivo
+
+```
+You are a senior competitive intelligence analyst for {industry}. Your goal is to
+produce a decision-grade competitive landscape report for the product "{our product}".
+
+Instructions:
+1. Use the Agent tool to dispatch 3 parallel investigators:
+   - Market landscape agent: map the top 5–8 competitors with core positioning
+   - Feature matrix agent: compile a feature-by-feature comparison using
+     publicly available sources (webSearch)
+   - Pricing & GTM agent: analyze pricing models, distribution channels, and
+     go-to-market motions
+
+2. Synthesize the three streams into a unified competitive report.
+
+3. Dispatch one review agent to stress-test the report: challenge any
+   assumption lacking a cited source, flag outdated data (>12 months), and
+   propose one "non-obvious" insight.
+
+4. Deliver the final report with the following sections:
+   - TL;DR (3 bullets)
+   - Competitor positioning map
+   - Feature matrix (markdown table)
+   - Pricing & GTM table
+   - Top 3 strategic implications for our product
+   - Caveats & data gaps
+```
+
+Em comparação com o Research Expert original: reduzido para 4 etapas, subfunções reduzidas de 6 para 3, lista de entregáveis totalmente reescrita como "seções do relatório".

package/concepts/pt-BR/UltraPlan.md

@@ -53,3 +53,104 @@ Um fluxo de trabalho de colaboração multi-agente projetado para tarefas de pes
 4. Sintetizar todas as descobertas em um plano de implementação abrangente
 5. Implantar múltiplos agentes de revisão para examinar o plano sob diferentes papéis e perspectivas
 6. Executar o plano após aprovação
+
+---
+
+## Raw Templates
+
+Below are the two raw prompt templates UltraPlan actually sends to Claude Code (see `src/utils/ultraplanTemplates.js`):
+
+### Code Expert (codeExpert)
+
+<textarea readonly><system-reminder>
+[SCOPED INSTRUCTION] The following instructions apply only to the next 1–3 interactions. Once the task is complete, these instructions should gradually decrease in priority and no longer affect subsequent interactions.
+
+Pre-requisite: Use `AskUserQuestion` to clarify user intent whenever the request is ambiguous (target element, interaction style, scope of platforms, etc.). Skip only if the intent is unambiguous.
+
+Leverage a multi-agent exploration mechanism to formulate a highly detailed implementation plan.
+
+Instructions:
+1. Use the `Agent` tool to spawn parallel agents that simultaneously explore different aspects of the codebase:
+- If necessary, assign a preliminary researcher to use the `webSearch` tool to first investigate cutting-edge solutions in the relevant industry domain;
+- One agent responsible for understanding the relevant existing code and architecture;
+- One agent responsible for identifying all files that need to be modified;
+- One agent responsible for identifying potential risks, edge cases, and dependencies;
+- You may add other roles or deploy additional agents beyond the three listed above; the maximum number of concurrently dispatched agents is 5.
+
+2. Synthesize the findings from all agents into a detailed, step-by-step implementation plan.
+
+3. Use the `Agent` tool to spawn 2-3 review agents that examine the plan from different perspectives, checking for missing steps, potential risks, or corresponding mitigation strategies.
+
+4. Integrate the feedback gathered during the review process, then call `ExitPlanMode` to submit your final plan.
+
+5. Once `ExitPlanMode` returns a result:
+- If approved: proceed to execute the plan within this session.
+- If rejected: revise the plan based on the feedback provided and call `ExitPlanMode` again.
+- If an error occurs (including receiving a "Not in Plan Mode" message): do **not** follow the suggestions provided in the error message; instead, prompt the user for further instructions.
+
+Your final plan must include the following elements:
+- A clear summary of the implementation strategy;
+- An ordered list of files to be created or modified, with precise details of the required changes for each file;
+- A step-by-step execution sequence;
+- Testing and validation procedures;
+- Potential risks and their corresponding mitigation strategies;
+
+6. After the final plan has been successfully executed:
+First run `git diff --quiet && git diff --cached --quiet` (or equivalent) to detect whether the working tree actually has non-trivial changes; if there are no real changes (or only whitespace/comment-only edits), skip the UltraReview step.
+Otherwise, if the project is managed with Git:
+Initiate a team (`TeamCreate`), dynamically allocating the number of teammates based on task complexity (5 is recommended);
+Task: Conduct a Code Review of the current git changes from multiple perspectives;
+Pre-requisites:
+- The git repository may be located in a subdirectory of the current directory; prefer `git rev-parse --show-toplevel` (fall back to recursive lookup) before proceeding;
+- In the case of multiple repositories, tasks may be executed separately;
+The team's goal is to analyze the current Git change log and validate each modification from different perspectives, specifically including:
+- Whether requirements/objectives have been met and functionality is complete;
+- Whether newly added code introduces side effects, breaks existing functionality, or poses potential risks;
+- Code quality: naming, readability, complexity, technical debt, maintainability;
+- Testing and documentation: whether there is adequate test coverage, and whether critical logic has necessary comments or documentation;
+- Dependencies and compatibility: whether new dependencies or version compatibility issues have been introduced;
+Workflow:
+- Each teammate, according to their own role, covers the review dimensions one by one and independently outputs a report;
+- After consolidating the reports, perform a cross-review to identify conflicts or shared concerns;
+- Distill specific, actionable modification suggestions and annotate them with priority levels (P0/P1/P2/P3);
+- Upon completion, adopt P0 items, and selectively adopt P1 items when they are concrete and low-risk; defer P2/P3 to backlog;
+- After execution is complete, close the team (`TeamDelete`);
+</system-reminder></textarea>
+
+### Research Expert (researchExpert)
+
+<textarea readonly><system-reminder>
+[SCOPED INSTRUCTION] The following instructions are intended for the next 1–3 interactions. Once the task is complete, these instructions should be gradually deprioritized and no longer influence subsequent interactions.
+
+Pre-requisite: Use `AskUserQuestion` to clarify the research scope, target audience, and deliverable format whenever the user's intent is ambiguous. Skip only if the intent is unambiguous.
+
+Leverage a multi-agent exploration mechanism to formulate an exceptionally detailed implementation plan.
+
+Instructions:
+1. Utilize the Agent tool to spawn parallel agents that simultaneously explore various facets of the requirements:
+- If necessary, deploy a preliminary investigator to conduct an initial survey of industry-specific solutions using `webSearch`;
+- If necessary, deploy a specialized investigator to research authoritative sources—such as academic papers, news articles, and research reports—using `webSearch`;
+- Assign an agent to synthesize the target solution, while simultaneously verifying the rigor and credibility of the gathered papers, news, and research reports;
+- If necessary, assign an agent to analyze competitor data to provide supplementary analytical perspectives;
+- If necessary, assign an agent to handle the implementation of a product demo (generating outputs such as HTML, Markdown, etc.);
+- If the task is sufficiently complex, you may assign additional teammates to the roles defined above, or introduce other specialized roles; you are permitted to schedule up to 5 teammates concurrently.
+
+2. Synthesize the findings from the aforementioned agents into a comprehensive, step-by-step implementation plan.
+
+3. Utilize the Agent tool to spawn a set of parallel review agents; these agents shall scrutinize the plan from multiple roles and perspectives to identify any omitted steps and to propose reasonable additions or optimizations.
+
+4. Consolidate the feedback received from the review agents, then invoke `ExitPlanMode` to submit your final plan.
+
+5. Upon receiving the result from `ExitPlanMode`:
+- If Approved: Proceed to execute the plan within this current session.
+- If Rejected: Revise the plan based on the provided feedback, and then invoke `ExitPlanMode` once again.
+- If an Error Occurs (including the message "Not in Plan Mode"): Do *not* follow the suggestions provided by the error message; instead, prompt the user for further instructions.
+
+Your final plan must include the following elements:
+- A clear summary of the proposed implementation strategy;
+- An ordered list of files to be created or modified, specifying the exact changes required for each;
+- A step-by-step sequence for executing the implementation;
+- Identification of potential risks and corresponding mitigation strategies;
+- Creative ideation and suggestions for advanced enhancements;
+- If a product demo was generated, place the corresponding demo output in an appropriate location and notify the user.
+</system-reminder></textarea>

package/concepts/ru/CustomUltraplanExpert.md

@@ -0,0 +1,165 @@
+# Custom UltraPlan Expert — Руководство по созданию
+
+## Что делают два поля ввода
+
+- **Имя эксперта**: подпись, отображаемая на кнопке роли в строке вариантов UltraPlan (макс. 30 символов). Это просто отображаемое имя, и оно **никогда** не отправляется в Claude Code.
+- **Тело промпта**: ваша инструкция роли. При отправке cc-viewer **автоматически** оборачивает её в теги `<system-reminder>...</system-reminder>` с заголовком области `[SCOPED INSTRUCTION]`. Поэтому **пишите только тело** — не добавляйте теги `<system-reminder>` самостоятельно.
+
+---
+
+## Как выглядит шаблон эксперта?
+
+Каждый встроенный эксперт (Code Expert / Research Expert) — это, по сути, блок `<system-reminder>`, внедряемый в контекст Claude Code. Ваш пользовательский эксперт проходит через тот же самый конвейер. Вот разбор шаблона **Research Expert**:
+
+```xml
+<system-reminder>
+[SCOPED INSTRUCTION] The following instructions are intended for the next 1–3
+interactions. Once the task is complete, these instructions should be gradually
+deprioritized and no longer influence subsequent interactions.
+
+Pre-requisite: Use `AskUserQuestion` to clarify the research scope, target
+audience, and deliverable format whenever the user's intent is ambiguous. Skip
+only if the intent is unambiguous.
+
+Leverage a multi-agent exploration mechanism to formulate an exceptionally
+detailed implementation plan.
+
+Instructions:
+1. Utilize the Agent tool to spawn parallel agents that simultaneously explore
+   various facets of the requirements:
+   - If necessary, deploy a preliminary investigator to conduct an initial
+     survey of industry-specific solutions using `webSearch`;
+   - If necessary, deploy a specialized investigator to research authoritative
+     sources—such as academic papers, news articles, and research reports—
+     using `webSearch`;
+   - Assign an agent to synthesize the target solution, while simultaneously
+     verifying the rigor and credibility of the gathered papers, news, and
+     research reports;
+   - If necessary, assign an agent to analyze competitor data to provide
+     supplementary analytical perspectives;
+   - If necessary, assign an agent to handle the implementation of a product
+     demo (generating outputs such as HTML, Markdown, etc.);
+   - If the task is sufficiently complex, you may assign additional teammates
+     to the roles defined above, or introduce other specialized roles; you are
+     permitted to schedule up to 5 teammates concurrently.
+
+2. Synthesize the findings from the aforementioned agents into a comprehensive,
+   step-by-step implementation plan.
+
+3. Utilize the Agent tool to spawn a set of parallel review agents; these
+   agents shall scrutinize the plan from multiple roles and perspectives to
+   identify any omitted steps and to propose reasonable additions or
+   optimizations.
+
+4. Consolidate the feedback received from the review agents, then invoke
+   `ExitPlanMode` to submit your final plan.
+
+5. Upon receiving the result from `ExitPlanMode`:
+   - If Approved: Proceed to execute the plan within this current session.
+   - If Rejected: Revise the plan based on the provided feedback, and then
+     invoke `ExitPlanMode` once again.
+   - If an Error Occurs: Do *not* follow the suggestions; prompt the user for
+     further instructions.
+
+Your final plan must include the following elements:
+- A clear summary of the proposed implementation strategy;
+- An ordered list of files to be created or modified, specifying the exact
+  changes required for each;
+- A step-by-step sequence for executing the implementation;
+- Identification of potential risks and corresponding mitigation strategies;
+- Creative ideation and suggestions for advanced enhancements;
+- If a product demo was generated, place the corresponding demo output in an
+  appropriate location and notify the user.
+</system-reminder>
+```
+
+---
+
+## Разбор по разделам
+
+### 1. Заголовок области `[SCOPED INSTRUCTION]` (обёртка — генерируется автоматически)
+> The following instructions are intended for the next 1–3 interactions...
+
+Это говорит Claude Code: **эти инструкции активны только для следующих 1–3 ходов**, затем затухают. Предотвращает «утечку» «персоны эксперта» в последующие несвязанные диалоги.
+
+**Эта строка генерируется cc-viewer автоматически. Вам не нужно её писать.**
+
+### 2. Начальное определение задачи (**это то, что вы должны переписать**)
+> Leverage a multi-agent exploration mechanism to formulate an exceptionally detailed implementation plan.
+
+Это «подлежащее-сказуемое-дополнение» всего шаблона: **оно сообщает Claude Code позицию и цель**. Дефолтная «многоагентная разведка + план реализации» хорошо подходит для задач **программной инженерии / планирования**, но выглядит неуклюже для многих других областей (рецензирование контента, анализ данных, копирайтинг, маркетинговые исследования, аудит соответствия…).
+
+**Мы настоятельно рекомендуем переписать эту строку под вашу цель**, например:
+
+- **Рецензент контента**: "You are a senior content reviewer specializing in {domain}. Your goal is to identify factual inaccuracies, tone inconsistencies, and structural weaknesses in the provided material."
+- **Конкурентный аналитик**: "Conduct a rigorous competitive analysis for {product category}. Produce a comparison matrix, positioning insights, and strategic recommendations."
+- **Копирайтер**: "Generate multiple creative copy variants for {scenario}, each with distinct positioning, tone, and call-to-action strategy."
+
+### 3. Шаги рабочего процесса (1–5 пунктов — **сократите или расширьте в зависимости от сложности**)
+
+Research Expert содержит 5 шагов: **исследование → синтез → ревью → отправка плана → выполнение**. Это обеспечивает «параллельная многоагентная работа + перекрёстное ревью + утверждение плана» — три уровня строгости, подходящие для задач высокой ставки/широкого охвата, но **избыточные для лёгких**.
+
+- **Простая задача** (одиночный поиск / маленькая правка): откажитесь от диспетчеризации многоагентности и ревью; просто «выдайте ответ» в один шаг.
+- **Умеренная задача**: оставьте «исследование → синтез → ревью»; уберите танец с ExitPlanMode; выдайте результат напрямую.
+- **Сложная, дорогостоящая задача** (крупный рефакторинг, сравнение нескольких вариантов, кросс-доменное исследование): сохраните все 5 шагов, возможно добавьте шаг «модель рисков» или «матрица сравнения вариантов».
+
+### 4. Подроли в шаге 1 (**адаптируйте под вашу область**)
+
+Research Expert перечисляет 6 потенциальных ролей (отраслевой разведчик, академический исследователь, синтезатор + проверяющий факты, аналитик конкурентов, продюсер демо, расширяемый слот). **Перепишите этот список под ваш сценарий**:
+
+- **Письмо**: "source collector + style analyst + fact checker"
+- **Анализ данных**: "data-cleaning agent + statistical modeling agent + visualization agent"
+- **Аудит кода**: "static-analysis agent + dependency-chain auditor + threat modeler"
+
+### 5. Контрольный список итогового результата (**согласуйте с вашей реальной потребностью**)
+
+> Your final plan must include the following elements: ...
+
+Оригинальный шаблон перечисляет 6 элементов «плана реализации». Ваш итоговый результат может быть совершенно другим:
+
+- **Исследовательский отчёт** → "Executive summary / Methodology / Key findings / Limitations / Action recommendations"
+- **Отчёт о ревью** → "Issue list / Severity rating / Fix suggestions / Before-after examples"
+- **Матрица сравнения** → "Dimension definitions / Scoring rubric / Conclusions / Recommendation rationale"
+
+---
+
+## Советы по созданию (TL;DR)
+
+1. **Сохраните обёртку**: строка `<system-reminder>` + `[SCOPED INSTRUCTION]` добавляется cc-viewer — не повторяйте.
+2. **Перепишите вступительное предложение**: укажите роль, цель и формат вывода в одной строке.
+3. **Гибко используйте рабочий процесс**: 1–2 шага для лёгких задач, полный 5-шаговый цикл — только для сложных.
+4. **Перепишите подроли шага 1**: дефолты (академические статьи / конкуренты / демо), вероятно, не то, что вам нужно.
+5. **Финальный «контрольный список результата» — это ваша планка качества**: пропишите структуру вывода — Claude Code будет строго ей следовать.
+
+---
+
+## Рефакторенный пример: Competitive Analyst
+
+```
+You are a senior competitive intelligence analyst for {industry}. Your goal is to
+produce a decision-grade competitive landscape report for the product "{our product}".
+
+Instructions:
+1. Use the Agent tool to dispatch 3 parallel investigators:
+   - Market landscape agent: map the top 5–8 competitors with core positioning
+   - Feature matrix agent: compile a feature-by-feature comparison using
+     publicly available sources (webSearch)
+   - Pricing & GTM agent: analyze pricing models, distribution channels, and
+     go-to-market motions
+
+2. Synthesize the three streams into a unified competitive report.
+
+3. Dispatch one review agent to stress-test the report: challenge any
+   assumption lacking a cited source, flag outdated data (>12 months), and
+   propose one "non-obvious" insight.
+
+4. Deliver the final report with the following sections:
+   - TL;DR (3 bullets)
+   - Competitor positioning map
+   - Feature matrix (markdown table)
+   - Pricing & GTM table
+   - Top 3 strategic implications for our product
+   - Caveats & data gaps
+```
+
+По сравнению с оригинальным Research Expert: сокращено до 4 шагов, подроли уменьшены с 6 до 3, список результатов полностью переписан как «разделы отчёта».

package/concepts/ru/UltraPlan.md

@@ -53,3 +53,104 @@ UltraPlan предлагает две экспертные роли для ра
 4. Синтез всех находок в комплексный план реализации
 5. Развёртывание нескольких агентов-рецензентов для проверки плана с разных ролей и перспектив
 6. Выполнение плана после утверждения
+
+---
+
+## Raw Templates
+
+Below are the two raw prompt templates UltraPlan actually sends to Claude Code (see `src/utils/ultraplanTemplates.js`):
+
+### Code Expert (codeExpert)
+
+<textarea readonly><system-reminder>
+[SCOPED INSTRUCTION] The following instructions apply only to the next 1–3 interactions. Once the task is complete, these instructions should gradually decrease in priority and no longer affect subsequent interactions.
+
+Pre-requisite: Use `AskUserQuestion` to clarify user intent whenever the request is ambiguous (target element, interaction style, scope of platforms, etc.). Skip only if the intent is unambiguous.
+
+Leverage a multi-agent exploration mechanism to formulate a highly detailed implementation plan.
+
+Instructions:
+1. Use the `Agent` tool to spawn parallel agents that simultaneously explore different aspects of the codebase:
+- If necessary, assign a preliminary researcher to use the `webSearch` tool to first investigate cutting-edge solutions in the relevant industry domain;
+- One agent responsible for understanding the relevant existing code and architecture;
+- One agent responsible for identifying all files that need to be modified;
+- One agent responsible for identifying potential risks, edge cases, and dependencies;
+- You may add other roles or deploy additional agents beyond the three listed above; the maximum number of concurrently dispatched agents is 5.
+
+2. Synthesize the findings from all agents into a detailed, step-by-step implementation plan.
+
+3. Use the `Agent` tool to spawn 2-3 review agents that examine the plan from different perspectives, checking for missing steps, potential risks, or corresponding mitigation strategies.
+
+4. Integrate the feedback gathered during the review process, then call `ExitPlanMode` to submit your final plan.
+
+5. Once `ExitPlanMode` returns a result:
+- If approved: proceed to execute the plan within this session.
+- If rejected: revise the plan based on the feedback provided and call `ExitPlanMode` again.
+- If an error occurs (including receiving a "Not in Plan Mode" message): do **not** follow the suggestions provided in the error message; instead, prompt the user for further instructions.
+
+Your final plan must include the following elements:
+- A clear summary of the implementation strategy;
+- An ordered list of files to be created or modified, with precise details of the required changes for each file;
+- A step-by-step execution sequence;
+- Testing and validation procedures;
+- Potential risks and their corresponding mitigation strategies;
+
+6. After the final plan has been successfully executed:
+First run `git diff --quiet && git diff --cached --quiet` (or equivalent) to detect whether the working tree actually has non-trivial changes; if there are no real changes (or only whitespace/comment-only edits), skip the UltraReview step.
+Otherwise, if the project is managed with Git:
+Initiate a team (`TeamCreate`), dynamically allocating the number of teammates based on task complexity (5 is recommended);
+Task: Conduct a Code Review of the current git changes from multiple perspectives;
+Pre-requisites:
+- The git repository may be located in a subdirectory of the current directory; prefer `git rev-parse --show-toplevel` (fall back to recursive lookup) before proceeding;
+- In the case of multiple repositories, tasks may be executed separately;
+The team's goal is to analyze the current Git change log and validate each modification from different perspectives, specifically including:
+- Whether requirements/objectives have been met and functionality is complete;
+- Whether newly added code introduces side effects, breaks existing functionality, or poses potential risks;
+- Code quality: naming, readability, complexity, technical debt, maintainability;
+- Testing and documentation: whether there is adequate test coverage, and whether critical logic has necessary comments or documentation;
+- Dependencies and compatibility: whether new dependencies or version compatibility issues have been introduced;
+Workflow:
+- Each teammate, according to their own role, covers the review dimensions one by one and independently outputs a report;
+- After consolidating the reports, perform a cross-review to identify conflicts or shared concerns;
+- Distill specific, actionable modification suggestions and annotate them with priority levels (P0/P1/P2/P3);
+- Upon completion, adopt P0 items, and selectively adopt P1 items when they are concrete and low-risk; defer P2/P3 to backlog;
+- After execution is complete, close the team (`TeamDelete`);
+</system-reminder></textarea>
+
+### Research Expert (researchExpert)
+
+<textarea readonly><system-reminder>
+[SCOPED INSTRUCTION] The following instructions are intended for the next 1–3 interactions. Once the task is complete, these instructions should be gradually deprioritized and no longer influence subsequent interactions.
+
+Pre-requisite: Use `AskUserQuestion` to clarify the research scope, target audience, and deliverable format whenever the user's intent is ambiguous. Skip only if the intent is unambiguous.
+
+Leverage a multi-agent exploration mechanism to formulate an exceptionally detailed implementation plan.
+
+Instructions:
+1. Utilize the Agent tool to spawn parallel agents that simultaneously explore various facets of the requirements:
+- If necessary, deploy a preliminary investigator to conduct an initial survey of industry-specific solutions using `webSearch`;
+- If necessary, deploy a specialized investigator to research authoritative sources—such as academic papers, news articles, and research reports—using `webSearch`;
+- Assign an agent to synthesize the target solution, while simultaneously verifying the rigor and credibility of the gathered papers, news, and research reports;
+- If necessary, assign an agent to analyze competitor data to provide supplementary analytical perspectives;
+- If necessary, assign an agent to handle the implementation of a product demo (generating outputs such as HTML, Markdown, etc.);
+- If the task is sufficiently complex, you may assign additional teammates to the roles defined above, or introduce other specialized roles; you are permitted to schedule up to 5 teammates concurrently.
+
+2. Synthesize the findings from the aforementioned agents into a comprehensive, step-by-step implementation plan.
+
+3. Utilize the Agent tool to spawn a set of parallel review agents; these agents shall scrutinize the plan from multiple roles and perspectives to identify any omitted steps and to propose reasonable additions or optimizations.
+
+4. Consolidate the feedback received from the review agents, then invoke `ExitPlanMode` to submit your final plan.
+
+5. Upon receiving the result from `ExitPlanMode`:
+- If Approved: Proceed to execute the plan within this current session.
+- If Rejected: Revise the plan based on the provided feedback, and then invoke `ExitPlanMode` once again.
+- If an Error Occurs (including the message "Not in Plan Mode"): Do *not* follow the suggestions provided by the error message; instead, prompt the user for further instructions.
+
+Your final plan must include the following elements:
+- A clear summary of the proposed implementation strategy;
+- An ordered list of files to be created or modified, specifying the exact changes required for each;
+- A step-by-step sequence for executing the implementation;
+- Identification of potential risks and corresponding mitigation strategies;
+- Creative ideation and suggestions for advanced enhancements;
+- If a product demo was generated, place the corresponding demo output in an appropriate location and notify the user.
+</system-reminder></textarea>