jarvis-ai-framework 1.0.0

package/skills/prod-roadmap-report/commands/status.memory.md

@@ -0,0 +1,192 @@
+---
+description: Memory context where AIs can access and get context informations to remember data, names, patterns, so agents understands requests like a colleague would.
+argument-hint: <write important informations like team names, projects, path of files and folders and other informations to help LLMs remember contexts and don't your mind.>
+---
+# Memory
+
+Memory makes Claude your workplace collaborator - someone who speaks your internal language.
+
+This command will create or update the memory of this skill.
+
+
+## Instructions
+
+### 1. Check What Exists
+
+Check the working directory for:
+
+```
+CLAUDE.md                      ← About and full description hot memory
+.claude/status-report-memory/  ← Folder memory of this skill
+  glossary.md      ← Full decoder ring (everything)
+  shortcuts.md     ← Index, summarys and shortcuts to know fast where get right informations
+```
+
+### 2. Create What's Missing
+**If and `CLAUDE.md` and/or  `status-report-memory/` don't exist:** This is a fresh setup — begin the memory bootstrap workflow (see below). Place these in the current working directory.
+
+If the memory exist, read, understand and maintain in you context to use when needed in during user interaction.
+
+### 3. Bootstrap Memory (First Run Only)
+Only do this if `status-report-memory/` don't exist yet.
+
+Ask the user these questions **one section at a time**, allowing them to provide detailed answers:
+
+**Squads and teams names:**
+```
+What are the names of your squads/teams?
+- Please list all squad names exactly (or very similar) as they should appear in reports
+
+This will help me learn what kind projects and teams you work with.
+```
+
+```
+What are your main product and the initiatives your are working right now?
+- Please list all the projects and products you are reponsible, and the main initiatives your team are working on. Names exactly (or very similar) as they appear in the reports
+
+This will help me learn what kind projects and teams you work with.
+```
+
+**Files, notes and folders**
+If the user has not provided enough information yet, ask:
+
+```
+Where do you keep your specs or status reports? This could be:
+- A local file (e.g. transcripts, markdown notes, docx, txt)
+- An app (e.g. Obsidian, Notion, Jira, Asana)
+- Meeting transcripts files or directory
+- Projects/Product directory
+
+This will help me learn your workplace terms.
+```
+
+**Once you have access to this:**
+
+Analyze it for potential projects, initiatives and teams names:
+- Names that might be teams, squads or areas
+- Project or Products references or codenames
+- Initiatives, features or work items references
+- Acronyms or abbreviations
+- Internal terms or jargon
+- Tags, YAML properties patterns and relations
+
+**For each item, decode it interactively:**
+Example:
+```
+Project/Iniative: "We are late to deliver a way to show negative debit in user digital account. This is blocked by Credit Card team."
+
+I see some terms I want to make sure I understand:
+
+1. **Digital Account** - This is a Product or Project?
+2. **Show the Negative Debit** - Is an working in progress task?
+3. **Credit Card team** - is another team working with your team?
+```
+
+Continue through each information, asking only about terms you haven't already decoded.
+
+Gather data from available installed MCP sources:
+- **Chat:** Recent messages, channels, DMs
+- **Email:** Sent messages, recipients
+- **Documents:** Recent docs, collaborators
+- **Calendar:** Meetings, attendees
+
+Build a braindump of people, projects, and terms found. Present findings grouped by confidence:
+- **Ready to add** (high confidence) — offer to add directly
+- **Needs clarification** — ask the user
+- **Low frequency / unclear** — note for later
+### 4. Write Memory Files
+From everything gathered, create `pm-memory/glossary.md`: full decoder ring (acronyms, terms, nicknames, codenames)
+
+```markdown
+# Workspace Context
+
+> **Purpose**: This file provides AI agents with essential context about workspace structure, naming conventions, and organizational patterns for Status Report tasks.
+> **Last Updated**: YYYY-MM-DD
+
+---
+
+## Me
+[Name], [Role] on [Team].
+
+## Main Products/Projects 
+| Name | What |
+|------|------|
+| **[Codename]** | [description] |
+
+## Actual initiatives
+| Name | What |Product/Project|
+|------|------|------|
+| **[Codename]** | [description] |[product/project name]|
+
+## Other Squads/Teams
+
+| Name | About |
+|------|------|
+| **[team/squad name]** | [description] |
+
+## People
+| Who | Role | Squad/Team |
+|-----|------|------|
+| **[Nickname]** | [Full Name], [role] | [Squad/Team name] |
+
+
+## References and Assets
+{list all notes (with links), files, folders and other references and sources used to create this report and that can be consulted in the future.}
+
+```
+
+When possible (only if you have high confidence), put link to the sources in the name of things. 
+
+### creating or updating the shortcut.md
+The `status-report-memory/shortcut.md` file is an index, to Agents and AI can know where to get informations and contexts faster. They contain paths, links and informations of files and other references sources where agents and AI can understand where get informations to give right and well informed aswers.
+
+#### Rules to write and use shortcut.md
+- You need to just read and understand this file. Don't get or read the files, contents, links and other sources the file is connecting. Do this just when needed.
+- When user create new files of status report, insert in the Local Files section.
+- If the user asks to record some file, source reference or link to be accessed in the future, insert this in the right section in the `shortcut.md`
+
+The output of this file need to follow the template above:
+
+```markdown
+# Status Report Fast Memory
+
+This document is an index file to get hot informations in a way the Agents and AIs can be well informed and know where to get informations.
+
+## Local Files
+
+Local files that are important and can always be consulted, with a high update rate.
+
+| Filename | root path |about |
+|------|------|------|
+| [filename](path-to-the-file-name.md) | {full local path to the file} | {read the file and get a coerent and concise description max 250 chars.} |
+
+## Links
+
+Links of websites, sources and references.
+
+| URL | about |
+|------|------|
+| {link or url} | {If the user don't give more information about it, read the source and get the meta description when available. max 250 chars. } |
+
+
+## Other references
+
+List of other sources, references and topics.
+
+| Name | about |
+|------|------|
+| {name of the source, topic, subject} | {If the user don't give more information about it, read the source and try to get the meta description when available. max 250 chars. } |
+
+```
+
+## Notes
+
+- If memory is already initialized, this just follow the user commands
+- Names are critical — always capture how projects/products/initiatives are actually referred to
+- If a source isn't available, skip it and note the gap
+- Memory grows organically through natural conversation after bootstrap
+- **Interactive Flow**: Ask questions one section at a time, don't overwhelm
+- **Smart Defaults**: Suggest common patterns based on existing files if possible
+- **Validation**: After each section, summarize what you understood for confirmation
+- **Incremental**: User can say "skip" for optional sections
+- **Editable**: Make it clear the file can be manually edited anytime

package/skills/prod-roadmap-report/commands/status.roadmap.preview.md

@@ -0,0 +1,94 @@
+---
+name: prod-roadmap-preview
+description: >
+  Use esta habilidade sempre que alguém perguntar sobre roadmap de produto, status de entrega, status de sprint, atualizações de equipe ou relatórios de status.
+  Aciona quando os usuários perguntam coisas como: "qual o roadmap?", "quais são as entregas do trimestre?", "status do time X", "o que está sendo feito?", "quando vai sair feature Y?", "me mostra o roadmap", "status report do time", "próximas entregas", "roadmap do produto", "o que está planejado?", "report do time", "andamento dos projetos".
+  Essa habilidade se conecta ao Jira (Atlassian) e às páginas wiki internas para buscar dados ao vivo sobre o roteiro do produto e relatórios de status da equipe.
+---
+
+# Status Roadmap Preview
+
+Este skill responde perguntas sobre roadmap de produto e status reports dos times, buscando dados ao vivo no `$TASK_MANAGER` definido e outras fontes de dados fornecidas pelo usuário
+
+## Quando usar essa skill?
+
+Use este fluxo sempre que alguém perguntar sobre:
+- Roadmap de produto
+- Status de entrega
+- Status de sprint
+- Relatórios de status de produto e desenvolvimento
+- times de produtos
+- Entrega de projetos e priorização
+
+---
+
+## Regras
+
+- Se não tiver informações suficientes sobre o `$TASK_MANAGER` utilizado pelo usuário, pergunte qual é o `$TASK_MANAGER` e adapte a resposta de acordo.
+- Sempre utilize MCPs configurados especificos do `$TASK_MANAGER` utilizado para buscar informações atualizadas
+- Procure na pasta de trabalho para entender mais sobre o projeto.
+- Caso não tiver informações sobre o projeto, fontes de dados ou `$TASK_MANAGER`, informe ao usuário que não tem informações suficientes e peça para ele fornecer mais detalhes.
+
+### Formato de Resposta padrão - Tabela por Mês
+
+A resposta padrão quando usuário pedir roadmap, deve exibir **mês atual** e os **próximos três meses**.
+Se o usuário pedir outro mês específico, mostre apenas aquele mês específico.
+
+Formato deve ser:
+```
+## {Mês atual}
+
+| Key                                                        | Summary           | Status           | Priority   | Due Date   | Resolvido |
+| ---------------------------------------------------------- | ----------------- | ---------------- | ---------- | ---------- | --------- |
+| [key-XXXX](https://url.of.task.manager.task/) | Summary | Status | Prioridade | 31/03/2026 | —         |
+| [key-XXXX](https://url.of.task.manager.task/) | Summary | Status | Prioridade | 31/03/2026 | —         |
+| [key-XXXX](https://url.of.task.manager.task/) | Summary | Status | Prioridade | 31/03/2026 | —         |
+
+## {nome próximo mês}
+
+| Key                                                        | Summary           | Status           | Priority   | Due Date   | Resolvido |
+| ---------------------------------------------------------- | ----------------- | ---------------- | ---------- | ---------- | --------- |
+| [key-XXXX](https://url.of.task.manager.task/) | Summary | Status | Prioridade | DD/MM/YYYY | —         |
+| [key-XXXX](https://url.of.task.manager.task/) | Summary | Status | Prioridade | DD/MM/YYYY | —         |
+| [key-XXXX](https://url.of.task.manager.task/) | Summary | Status | Prioridade | DD/MM/YYYY | —         |
+
+
+## {nome próximo mês 2}
+
+| Key                                                        | Summary           | Status           | Priority   | Due Date   | Resolvido |
+| ---------------------------------------------------------- | ----------------- | ---------------- | ---------- | ---------- | --------- |
+| [key-XXXX](https://url.of.task.manager.task/) | Summary | Status | Prioridade | DD/MM/YYYY | —         |
+| [key-XXXX](https://url.of.task.manager.task/) | Summary | Status | Prioridade | DD/MM/YYYY | —         |
+| [key-XXXX](https://url.of.task.manager.task/) | Summary | Status | Prioridade | DD/MM/YYYY | —         |
+
+## {nome próximo mês 3}
+
+| Key                                                        | Summary           | Status           | Priority   | Due Date   | Resolvido |
+| ---------------------------------------------------------- | ----------------- | ---------------- | ---------- | ---------- | --------- |
+| [key-XXXX](https://url.of.task.manager.task/) | Summary | Status | Prioridade | DD/MM/YYYY | —         |
+| [key-XXXX](https://url.of.task.manager.task/) | Summary | Status | Prioridade | DD/MM/YYYY | —         |
+| [key-XXXX](https://url.of.task.manager.task/) | Summary | Status | Prioridade | DD/MM/YYYY | —         |
+
+...
+```
+
+---
+
+## Fluxo 2: Status dos projetos de times
+
+### Quando usar
+Sempre que o usuário pedir: status report, atualização do time, o que o time está fazendo, últimas atualizações, etc.
+
+### Passo 1 — Perguntar qual time
+Se o usuário não especificou o time, pergunte:
+
+> "Qual time você quer ter visibilidade?"
+
+
+## Comportamento Geral
+
+- Sempre responda em **português**
+- Seja objetivo e use formatação em tabela quando possível
+- Se não encontrar dados, informe claramente qual fonte não retornou resultados
+- Datas devem ser exibidas no formato **DD/MM/AAAA**
+- Quando não houver data de resolução, exiba **—**

package/skills/prod-roadmap-report/references/detailed-guide.md

@@ -0,0 +1,236 @@
+# Project Status Report - Detailed Guide
+
+This file contains detailed examples, section guidelines, and edge cases for the project status report skill.
+
+## Section Guidelines
+
+### Summary Paragraph
+- Write in prose, not bullet points
+- Max 200 words, divided in 2-4 paragraphs typical. Keep it concise but complete
+- Capture the main discussion topics, based on time discussed and how many times was cited in the discussed
+- Include context and key points
+- Use neutral, factual language
+
+### On Track
+- Items showing positive progress
+- Completed milestones
+- Things proceeding as planned
+- Format: **Bold topic name**: Description
+- If nothing is explicitly on track, use empty list or omit section
+- ignore this block in the final output if empty or don't have content
+
+### Problems
+
+- Current blockers
+- Active issues
+- Things not working as expected
+- Format: **Bold problem name**: Description
+- Only include explicitly mentioned problems
+- ignore this block in the final output if empty or don't have content
+
+### Risks & Concerns
+
+- Potential future problems
+- Uncertainties
+- Dependencies that might fail
+- When the topic is late or will re-scheduled
+- Format: **Bold concern/risk name**: Description
+- Only include explicitly mentioned concerns
+- ignore this block in the final output if empty or don't have content
+
+### Actions
+
+- Specific next steps
+- Assigned tasks (include assignee if mentioned)
+- Decisions that require follow-up
+- Simple list format
+- Only include explicitly stated actions
+- Don't invent actions
+
+## Historical Context Usage
+
+If previous reports or knowledge base information is available:
+
+- **Team name consistency**: Use established team names from previous reports
+- **Project naming**: Follow naming conventions from historical reports
+- **Recurring issues tracking**: Check if current problems were mentioned in previous reports
+- **Action item follow-up**: Cross-reference with previous action items to track completion
+- **Pattern recognition**: Identify recurring risks or concerns
+- **Context enrichment**: Use historical context to better understand abbreviations, codenames, or internal references
+
+**Important**: Historical context is used ONLY for:
+- Understanding terminology and naming patterns
+- Maintaining consistency with previous reports
+- Recognizing recurring themes
+
+**NEVER use historical context for:**
+- Inventing current status updates
+- Filling in missing information
+- Making assumptions about current progress
+
+## Update Workflow
+
+When updating an existing report:
+
+1. **Read original report carefully**: Note language, section structure, formatting choices
+2. **Preserve exact structure**: Keep the same headers, section order, and formatting style
+3. **Insert new content BEFORE old content**: Maintain reverse chronological order
+4. **Use current date**: DD-MM-YYYY format for new entry
+5. **Maintain language consistency**: Use same language as original report
+
+**Example of correct update structure:**
+
+**Existing report on 01-12-2025:**
+
+```markdown
+## Backend Team
+### Payment API
+
+#### 01-12-2025
+Database migration completed successfully. All endpoints tested and working.
+
+**On track:**
+- **Database migration**: Completed without downtime
+...
+```
+
+**New update on 02-12-2025:**
+
+```markdown
+## Backend Team
+### Payment API
+
+#### 02-12-2025
+Performance optimization implemented. Response times improved by 40%.
+
+**On track:**
+- **API performance**: Response time reduced from 500ms to 300ms
+...
+
+#### 01-12-2025
+Database migration completed successfully. All endpoints tested and working.
+
+**On track:**
+- **Database migration**: Completed without downtime
+...
+```
+
+New content is inserted BEFORE the old content, maintaining reverse chronological order.
+
+## Handling Edge Cases
+
+### Single Project, No Team Name
+
+```markdown
+## [Team Name To Be Defined]
+### {Project Name}
+#### {date}
+{content}
+```
+
+### Multiple Teams, Same Project
+
+Create separate sections for each team's perspective on the project.
+
+### No Clear Project Structure
+
+If the input is just general updates without clear project boundaries:
+
+1. Ask the user (in their language) if the update refers to a specific project or is a general team status
+2. Structure based on user response
+
+### Conflicting Information
+
+When there's conflicting information about status, dates, or other critical details:
+
+- Stop and ask the user (in their language) to clarify which information is correct, quoting both conflicting statements
+
+## Example Interactions
+
+### Example 1: Well-Structured Input
+
+**User provides:** Meeting transcription with clear team names, project discussions, problems, and actions.
+
+**Assistant should:**
+
+1. Extract all information
+2. Structure into template
+3. Generate complete markdown report
+4. No questions needed
+
+### Example 2: Ambiguous Input
+
+**User provides:** Slack messages with technical discussions but no clear team attribution.
+
+**Assistant should:**
+
+1. Extract technical content
+2. Organize into template structure
+3. Ask (in user's language) which team is responsible for this work
+4. Generate report after clarification
+
+### Example 3: Multiple Projects
+
+**User provides:** Long transcription covering 3 different initiatives.
+
+**Assistant should:**
+
+1. Identify the 3 initiatives
+2. Extract information for each
+3. Create single markdown with 3 sections (one per initiative)
+4. Follow template for each section
+
+### Example 4: Update Existing Report
+
+**User provides:** Previous report file + new meeting notes
+
+**Assistant should:**
+
+1. Read and analyze previous report structure (language, sections, format)
+2. Extract new information from meeting notes
+3. Merge new info into existing structure while preserving EXACT format
+4. Update date to current date
+5. Insert new content BEFORE old content (reverse chronological)
+6. Return updated report maintaining all original structural choices
+
+### Example 5: Update with Historical Context
+
+**User provides:** "Update the Backend Team report with today's updates" + new information
+
+**Assistant should:**
+
+1. Search for previous Backend Team reports in conversation history or knowledge base
+2. Use historical context to understand team structure, naming conventions
+3. Follow established patterns from previous reports
+4. Generate updated report consistent with team's reporting style
+5. Insert new update block before existing blocks
+
+## Multiple Projects Structure Example
+
+**Correct:**
+
+```markdown
+## Alpha Team
+
+### Dashboard Analytics Project
+#### 01-12-2025
+{summary}
+**On track:**
+...
+
+### Mobile App Redesign
+#### 01-12-2025
+{summary}
+**On track:**
+...
+
+## Beta Team
+
+### Social Login Feature
+#### 01-12-2025
+{summary}
+**On track:**
+...
+```
+
+Note: Each team header (`##`) appears only once, with all their projects grouped under it as subheaders (`###`).