Menu
Fetch a GitHub issue from bbvch-ai/aihub-core and create an implementation plan using monorepo scope analysis, ADR checks, and scaffold skill mapping. Use when user says 'plan this issue', 'how should I implement issue
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Baseado na classificação ocupacional SOC
| name | plan-issue |
| description | Fetch a GitHub issue from bbvch-ai/aihub-core and create an implementation plan using monorepo scope analysis, ADR checks, and scaffold skill mapping. Use when user says 'plan this issue', 'how should I implement issue |
| allowed-tools | Bash, Read, Grep, Glob |
Fetch issue $ARGUMENTS from bbvch-ai/aihub-core and produce a scoped implementation plan.
Use the GitHub MCP server to gather structured issue data:
mcp__github__issue_read with method: "get" — title, body, author, assignees, milestonemcp__github__issue_read with method: "get_comments" — clarifications, decisions, contextmcp__github__issue_read with method: "get_sub_issues" — task breakdown if presentmcp__github__issue_read with method: "get_labels" — categorization and priorityAll calls use owner: "bbvch-ai", repo: "swiss-ai-hub", issue_number: $ISSUE_NUMBER.
Optionally check the project board for priority and status context:
gh project item-list 13 --owner bbvch-ai --format json | jq '.items[] | select(.content.number == $ISSUE_NUMBER)'
Extract from the issue: Goal (what needs to be achieved), Constraints (requirements or limitations), Context (related issues, PRs, or decisions).
Map the issue requirements to monorepo scopes:
| Scope | Responsibility |
|---|---|
packages/core | Shared library (events, entities, NATS, auth, config) |
packages/api | REST API + WebSocket gateway (FastAPI controllers, services) |
packages/agent | AI agent definitions and workflows (LlamaIndex) |
packages/pipeline | Data ingestion pipelines (Dagster assets, resources) |
packages/process | Business process orchestration (agent + human + program steps) |
packages/bot | Collaboration platform integrations (MS Teams, Slack) |
packages/web | Frontend admin UI (Nuxt 3, Vue 3, PrimeVue) |
For each affected scope:
{scope}/CLAUDE.md to understand its architecture and patternsGrep and Globls docs/arc42/decisions/
Read any ADR that covers related technology choices. Flag if the implementation approach may need a new ADR (new dependency, new framework, fundamental pattern change).
Check if scaffold skills exist for the type of work needed:
/scaffold-agent/scaffold-pipeline/scaffold-process/scaffold-api-endpoint + /scaffold-api-service + /scaffold-api-repository/scaffold-frontend-page + /scaffold-composable/scaffold-bot-handler/validate-events + /nats-eventsReference these in the plan so the implementer can invoke them.
Structure the plan with these sections:
packages/core changes, list downstream scopes that need testingpackages/core/swiss_ai_hub/core/events/make test, new test files neededpackages/core before downstream scopes)packages/core is modified, list all downstream scopes that import from it| Problem | Solution |
|---|---|
| Issue not found | Verify issue number: gh issue list -R bbvch-ai/aihub-core |
| Issue is vague | List what needs clarification, suggest asking the issue author |
| Spans too many scopes | Break into sub-tasks, suggest splitting the issue |
| No existing pattern to follow | Flag as risk — may need an ADR for the new pattern |
| Sub-issues exist | Use get_sub_issues to get the task breakdown |
Generate a new AI agent with all boilerplate (agent class, events, config with form duality, memory, LLM streaming, i18n, BDD tests, entry point). Includes pattern catalog, execution model reference, and implementation checklist. Use when user says "create new agent", "scaffold an agent", "generate agent boilerplate", "add AI agent", "new workflow agent", or "build an agent for X". Do NOT use for debugging agents (use /debug-agent), event infrastructure (use /nats-events), or process orchestration (use /scaffold-process).
Manage the Docker Compose development stack and understand the Jinja2 template-based compose generation system. Covers start/stop/health/logs/restart/status/ports for the dev stack, the template-to-generate-to-commit workflow, and documentation lookup via Context7 MCP. Use when user says 'start docker', 'stop containers', 'check service health', 'show logs', 'restart service', 'docker status', 'which ports are used', 'is the API running', 'regenerate compose', or 'how does docker compose generation work'. Pass action as argument. Do NOT use for reviewing Docker Compose changes (use deployment-reviewer agent) or debugging Dagster pipelines (use /debug-pipeline).
Author a new, convention-conformant GitHub issue for bbvch-ai/aihub-core — title format, the In scope / Out of scope / Accepted when body structure, area:* and version labels, sibling and blocked-by cross-references — then create it and add it to the AI-Scrum board (org project 37) with Item Type, Priority, and Status set. Use when user says 'write an issue', 'create a GitHub issue', 'open an issue for X', 'file a bug', 'draft an epic', or 'add a story to the board'. Do NOT use for planning an existing issue's implementation (use plan-issue), breaking an issue into sub-issues (use splice-issue), or creating a PR (use create-pr).
End-to-end guide for setting up a new bot connection in swiss_ai_hub.bot. Covers Azure App Registration, Bot Channels Registration, Teams/Slack channel config, PathEntity creation in bot_paths collection, and DevTunnel for local dev. Use when user says "set up a bot", "connect bot to Teams", "connect bot to Slack", "configure bot channel", "create bot connection", "DevTunnel setup", "bot local development", or "Azure bot registration". Do NOT use for bot handler code scaffolding (use scaffold-bot-handler), bot architecture questions (use bot-framework), or agent debugging (use debug-agent).
Use when working with `.c4`/`.likec4` files or LikeC4 CLI/config questions where exact DSL/CLI syntax is required, especially for strict command/snippet-first answers, validate/export flags, predicates `*`/`_`/`**`, deployment snippets, dynamic views, or relationship extension matching.
Run a comprehensive frontend code audit of the Nuxt 3 admin interface. Checks Pinia-Colada query/mutation patterns, TypeScript strictness, component architecture, SDK freshness, i18n coverage, and Tailwind usage. Use when user says 'audit the frontend', 'check frontend code quality', 'find unused components', 'composable health check', 'pinia-colada audit', 'typescript errors', or 'frontend code review'. Do NOT use for backend API audits (use scaffold-api-endpoint) or bot UI issues (use debug-bot). Reports issues with file locations and severity levels.