start-work

A workflow for starting work on a GitHub issue by creating a feature branch and preparing the project context. A GitHub issue is a tracked task or bug in a repository.

In plain words
What is it for?
Use it when beginning development on a new numbered GitHub issue.
Why use it?
It gathers the issue requirements and puts the work on a suitable branch before implementation begins.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/makigjuro/cloudstack-ai-plugins/start-work
Any agent
npx skills add makigjuro/cloudstack-ai-plugins --skill start-work
Clone the repo
git clone --depth 1 https://github.com/makigjuro/cloudstack-ai-plugins

Made for: Claude Code, Codex.

Per session 41 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 921 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What 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.

ModelPer sessionOnce invoked
Fable 5 $0.00041 $0.00921
Opus 5 $0.00020 $0.00461
Sonnet 5 $0.00008 $0.00184
Haiku 4.5 $0.00004 $0.00092

Measured 2d ago against content hash b41b2701c65a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

start-work 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.

plugins/dev-workflow/skills/start-work/SKILL.md · 102 lines

How it starts

The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Start Work

Create a feature branch from a GitHub issue and prepare the working context.

Configuration

Read cloudstack.json from the project root at the start of execution. Extract relevant fields:

  • REPO_OWNER / REPO_NAME = repository.owner / repository.name (default: detect from git remote -v)

If cloudstack.json does not exist, auto-detect by parsing the GitHub remote URL:

git remote get-url origin | sed -E 's|.*github\.com[:/]([^/]+)/([^/.]+).*|\1 \2|'

Arguments

  • {issue} -- GitHub issue number (required). Pass as argument: /start-work 42

Process

  1. Fetch issue details -- Read the issue title and body to understand the scope.
  2. Ensure clean working tree -- Check git status for uncommitted changes. Warn the user if the tree is dirty.
  3. Pull latest main -- Ensure the branch starts from the latest main.
  4. Create and checkout branch -- Use the naming convention below.
  5. Report -- Show the branch name and a summary of the issue tasks.

Branch Naming Convention

{username}/{issue-number}-{short-slug}

Derive {username} from the git config user.name (lowercase, no spaces). Derive {short-slug} from the issue title (lowercase, hyphens, max 50 chars, no special characters).

Examples:

  • jdoe/42-user-group-assignments
  • jdoe/15-order-versioning-api

Commands

Fetch the issue using MCP for structured data:

mcp__github-mcp-server__get_issue(owner: "{REPO_OWNER}", repo: "{REPO_NAME}", issue_number: {issue})

This returns structured JSON with title, body, labels, and state -- no CLI parsing needed.

# Check working tree
git status --short

# Update main and create branch
git fetch origin main
git checkout -b {branch-name} origin/main

After Starting -- Automatic Implementation

Once the branch is created, do not stop. Immediately continue with implementation:

  1. Analyze the issue -- Parse acceptance criteria, identify affected services, and determine which layers need changes (Domain, Application, Infrastructure, Endpoints, Frontend, Tests).
  2. Create a task list -- Use TodoWrite to track each implementation step derived from the issue.
  3. Implement -- Follow the layer order: Domain -> Application -> Infrastructure -> Endpoints -> Frontend -> Tests. Use the appropriate scaffolding skills (/add-entity, /add-command, /add-query, /add-event-handler, /add-migration, /add-feature) where they apply. Commit logically after each meaningful change.
  4. Verify as you go -- Run tests and architecture checks between steps. Fix issues before moving on.
  5. When done -- Run /complete-task to go through the full completion workflow (build, lint, tests, review, PR).

Read the full file on GitHub · 102 lines

Changes

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.

  1. 2d ago First seen · 102 lines · 41 tokens per session scan A b41b2701c65a

Subscribe to this mod's changes

start-work is a skill published in the GitHub repository makigjuro/cloudstack-ai-plugins (1 stars, last pushed 1mo ago), licensed MIT. It adds 41 tokens to every session and 921 once invoked, about $0.0002 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 2026-08-31.