start · ulpi-io — Claude Code skill: what it does, how to install
start is a Claude Code skill that routes incoming tasks, selects domain skills, evaluates sub-agent delegation, and limits context exploration before code edits.
no license 152 installs ★ 5 updated 2026-08-31
npx skills add ulpi-io/skills --skill start -a claude-codeWhat start does
This skill serves as an internal intake layer rather than a direct slash command. Because it sets user-invocable: false, the assistant runs it behind the scenes at the start of substantive tasks. It evaluates whether a domain skill already handles the issue, if an isolated agent is needed, what files to inspect first, and whether deep task tracking is necessary.
Instead of acting as a full planning system, it hands heavy dependency mapping off to plan-to-task-list-with-dag. Routine requests proceed directly to execution without creating tracking noise. Full capability inventories and file discovery are skipped unless requested directly.
The skill expects access to Read, Glob, Grep, Skill, and Agent. If an applicable domain skill is discovered during routing, the rules require immediate invocation rather than merely mentioning it in the output.
What it helps with
- route incoming engineering tasks to domain-specific skills immediately
- evaluate whether delegating to an isolated agent improves execution
- gather strictly necessary context files before editing code
- prevent excessive planning overhead on small direct edits
Example prompts for the agent
- Build an endpoint for exporting transaction reports to CSV
- Fix the race condition inside the background queue worker
How to install start
1Skills CLI (Vercel)
npx skills add ulpi-io/skills --skill start -a claude-code2Manually: put the files in place
~/.claude/skills/start/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/start/
1Skills CLI (Vercel)
npx skills add ulpi-io/skills --skill start -a codex2Manually: put the files in place
.agents/skills/start/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add ulpi-io/skills --skill start -a cursor2Manually: put the files in place
.cursor/skills/start/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install ulpi-io/skills start2Manually: put the files in place
.github/skills/start/SKILL.md1Skills CLI (Vercel)
npx skills add ulpi-io/skills --skill start -a gemini-cli2Platform CLI
gemini skills install https://github.com/ulpi-io/skills --path startManually: put the files in place
.devin/skills/start/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add ulpi-io/skills --skill start -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about start
How to install start in Claude Code?
Run npx skills add ulpi-io/skills --skill start -a claude-code in your terminal.
What does the start skill do?
It acts as an intake router that selects domain skills, evaluates agent delegation, and limits code inspection to what is strictly necessary.
Can you invoke start directly as a slash command?
No, it is configured with user-invocable: false and is designed strictly as an internal routing step when taking on new work.
Similar skills
All skills for Agent setup: memory, context, orchestration →start by other authors comes first.