create-issue
Create a well-structured task issue on GitHub that CI will validate and auto-sync to the project board. Use when user asks to create a task, file an issue, plan work, break down a feature, or make a new issue for any Phoenix repo.
Best use case
create-issue is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Create a well-structured task issue on GitHub that CI will validate and auto-sync to the project board. Use when user asks to create a task, file an issue, plan work, break down a feature, or make a new issue for any Phoenix repo.
Teams using create-issue should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/create-issue/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How create-issue Compares
| Feature / Agent | create-issue | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Create a well-structured task issue on GitHub that CI will validate and auto-sync to the project board. Use when user asks to create a task, file an issue, plan work, break down a feature, or make a new issue for any Phoenix repo.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Create task issue — Phoenix Agentic Engine Interface ## Overview Create task issues using the structured `task.yml` template so that the `issue-validate.yml` CI workflow can parse fields, validate them, add the issue to the project board, and auto-populate board fields (Status, Priority, Size, Area, Work mode, Lock Key, Needed Files, Depends On). ## Mandatory first step Check the project board to avoid duplicate work and understand current context: - Project board: https://github.com/users/rivie13/projects/3 - Check for existing issues covering the same scope before creating a new one: ``` mcp_github_github_search_issues(owner="rivie13", repo="Phoenix-Agentic-Engine-Interface", query="<keywords>") ``` ## Required fields Every task issue MUST include all of these in the issue body using the exact heading format below (this is what CI parses): | Field | Heading in body | Required | Example | |-------|----------------|----------|---------| | Target repo | `### Target repository` | Yes | `Phoenix-Agentic-Engine-Interface` | | Parent branch | `### Parent branch (PR target)` | Yes | `feature/phase-3-sdk-client` | | Working branch | `### Working branch` | Yes | `subfeature/task/sdk-client-auth` | | Work mode | `### Work mode` | Yes | `Local IDE` / `CLI Agent` / `Cloud Agent` | | Separate worktree | `### Separate worktree?` | Yes | `No` / `Yes` / `N/A (Cloud Agent)` | | Priority | `### Priority` | Yes | `P1 — High` | | Size | `### Size` | Yes | `M — Medium (half day to full day)` | | Area | `### Area` | Yes | `sdk/client` | | Description | `### Task description` | Yes | Free-form, detailed enough for agent | | Acceptance criteria | `### Acceptance criteria` | Yes | Checkboxes | | Key files | `### Key files` | No | File paths | | Depends on | `### Depends on` | No | `#42, #57` | | Lock key | `### Lock key` | No | `sdk/client/auth/*` | ## Valid values ### Priority - `P0 — Critical / blocker` - `P1 — High` - `P2 — Medium` - `P3 — Low` ### Size - `XS — Extra small (< 1 hour)` - `S — Small (1-4 hours)` - `M — Medium (half day to full day)` - `L — Large (multi-day)` - `XL — Extra large (multi-week)` ### Work mode - `Local IDE` — Human in VS Code - `CLI Agent` — Codex / Copilot CLI / other in a worktree - `Cloud Agent` — GitHub Copilot cloud coding agent ### Area (global — applies across all repos) `module/assistant-ui`, `module/mcp`, `module/agent`, `module/addons`, `core`, `gateway`, `worker`, `orchestrator`, `sdk/client`, `sdk/core`, `contracts`, `infra`, `app/pages`, `components`, `content`, `public`, `api`, `domain`, `tests`, `docs`, `ci` ### Branch naming convention - Feature branches: `feature/<description>` - Sub-feature / task: `subfeature/<type>/<description>` - Types: `task`, `fix`, `chore`, `docs` - Direct fixes: `fix/<description>`, `chore/<description>`, `docs/<description>` ## Creating an issue (step by step) ### 1. Determine execution context - **Which repo** will the work happen in? - **What parent branch** does the PR merge into? (Usually a `feature/*` branch, or `main`) - **What working branch** will be created? - **What work mode** — Local IDE, CLI Agent, or Cloud Agent? ### 2. Determine board fields - **Priority**: How urgent? P0 = blocker, P3 = nice-to-have - **Size**: How big? XS = trivial, L = multi-day - **Area**: What part of the codebase? ### 3. Write the issue body Format the body with exactly these markdown headings (CI parses them): ```markdown ### Target repository Phoenix-Agentic-Engine-Interface ### Parent branch (PR target) feature/phase-3-sdk-client ### Working branch subfeature/task/sdk-client-auth ### Work mode Local IDE ### Separate worktree? No ### Priority P2 — Medium ### Size S — Small (1-4 hours) ### Area sdk/client ### Task description Add authentication support to the SDK client... ### Acceptance criteria - [ ] Auth tokens are passed in headers - [ ] All tests pass - [ ] No lint errors ### Key files - sdk/client/auth.ts - tests/client/auth.test.ts ### Depends on _No response_ ### Lock key _No response_ ``` ### 4. Create via MCP tool ``` mcp_github_github_issue_write( method="create", owner="rivie13", repo="Phoenix-Agentic-Engine-Interface", title="[Task]: <concise title>", body="<body from step 3>", labels=["task"] ) ``` ### 5. What happens automatically CI will: 1. Parse all fields from the issue body headings. 2. Validate required fields are present and well-formed. 3. Add the issue to the project board (if not already there). 4. Set board fields: Status → Ready, Priority, Size, Area, Work mode. 5. Set text fields: Depends On, Lock Key, Needed Files (if provided). 6. Add a `base-branch:<branch>` label for cloud-agent resolution. 7. Comment with a success summary or validation errors. If validation fails, CI adds a `needs-triage` label and comments with specific errors. Edit the issue to fix them — CI re-validates on edit. ## Breaking down features into tasks When a feature is too large for a single issue: 1. Create the parent feature branch: `feature/<name>` 2. Break into sub-tasks, each as a separate issue 3. Each sub-task's **parent branch** = `feature/<name>` 4. Each sub-task's **working branch** = `subfeature/task/<name>` 5. Set `Depends on` to chain sequential tasks ## Related repos | Repo | GitHub | |------|--------| | Engine | `rivie13/Phoenix-Agentic-Engine` | | Backend | `rivie13/Phoenix-Agentic-Engine-Backend` | | Website Frontend | `rivie13/Phoenix-Agentic-Website-Frontend` | | Website Backend | `rivie13/Phoenix-Agentic-Website-Backend` | | Supervisor | `rivie13/Phoenix-Agentic-Workspace-Supervisor` | | Command Center | `rivie13/Phoenix-Agentic-VSCode-CommandCenter` | Use the same `task.yml` template when creating issues in any repo. The `create-issue` skill exists in all repos with identical guidance. ```
Related Skills
mcp-create-declarative-agent
Skill converted from mcp-create-declarative-agent.prompt.md
kitt-create-slash-commands
Expert guidance for creating slash commands. Use when working with slash commands, creating custom commands, understanding command structure, or learning YAML configuration.
kitt-create-plans
Create hierarchical project plans optimized for solo agentic development. Use when planning projects, phases, or tasks that the AI agent will execute. Produces agent-executable plans with verification criteria, not enterprise documentation. Handles briefs, roadmaps, phase plans, and context handoffs.
gh-issues
Fetch GitHub issues, spawn sub-agents to implement fixes and open PRs, then monitor and address PR review comments. Usage: /gh-issues [owner/repo] [--label bug] [--limit 5] [--milestone v1.0] [--assignee @me] [--fork user/repo] [--watch] [--interval 5] [--reviews-only] [--cron] [--dry-run] [--model glm-5] [--notify-channel -1002381931352]
create-workflow
Create Jazz workflow automation files (WORKFLOW.md). Use this for scheduling Jazz agents to run recurring tasks. For OS-level scripts/commands, use create-system-routine.
create-rule
Create persistent AI agent rules and instructions. Use when you want to create a rule, add coding standards, set up project conventions, configure file-specific patterns, or create AGENTS.md/GEMINI.md rule files across Cursor, Gemini CLI, or Codex.
create-prompt
Expert prompt engineering for creating effective prompts for Claude, GPT, and other LLMs. Use when writing system prompts, user prompts, few-shot examples, or optimizing existing prompts for better performance.
create-expert-skill
Create production-ready skills from expert knowledge. Extracts domain expertise and system ontologies, uses scripts for deterministic work, loads knowledge progressively. Use when building skills that must work reliably in production.
create-event-handlers
Sets up RabbitMQ event publishers and consumers following ModuleImplementationGuide.md Section 9. RabbitMQ only (no Azure Service Bus). Creates publishers with DomainEvent (tenantId preferred), consumers with handlers, naming {domain}.{entity}.{action}, required fields (id, type, version, timestamp, tenantId, source, data). Use when adding event-driven communication, async workflows, or integrating via events.
create-custom-prompt
Prompt for creating custom prompt files
create-assistant
Create and configure Vapi voice AI assistants with models, voices, transcribers, tools, hooks, and advanced settings. Use when building voice agents, phone bots, customer support assistants, or any conversational AI that handles phone or web calls.
create-agents-md
Create or rewrite AGENTS.md files for Open Mercato packages and modules. Use this skill when adding a new package, creating a new module, or when an existing AGENTS.md needs to be created or refactored. Ensures prescriptive tone, MUST rules, checklists, and consistent structure across all agent guidelines.