trainingsites/campus-ai-os/campus-first-win
Claude Code and agent skill by trainingsites/campus-ai-os
- community
Checking that the SKILL.md file is still in the repository.
Produce the owner's first real deliverable within 10 minutes of install. Picks the deliverable by delivery type from workspace.json - community prompts, a lesson plan, or a session prep pack - or offers the active pack's own domain skill when the pack ships one, and saves it to outputs/. Use when the owner says "/campus-first-win", "show me what you can do", or right after /campus-start.
Install
npx vskill@latest install trainingsites/campus-ai-os/campus-first-winSKILL.md
/campus-first-win - Real Work in 5 Minutes
The retention moment. The owner watches their new team produce actual, usable work before they've configured anything. One deliverable, done well, saved where they can find it.
Guard
Requires workspace.json (run /campus-start first if missing). Works fine with all shared/ files at defaults - use the founding answers - business.what, business.sells_today, business.audience (nested business object), and structure from workspace.json, plus the resolved identity context - as context. Any of the business fields may be empty on a quick-path campus that hasn't finished onboarding; fall back to the resolved identity context (resolved through CONTEXT_CANON, so it finds the file whatever the campus calls it) or ask one question. If they sell nothing yet, the first win is audience-building work (community prompts or a lead-magnet outline).
Step 1 - Pick by delivery type (from workspace.json)
| Delivery | Deliverable | |
|---|---|---|
| community | 5 ready-to-post discussion prompts for their topic and audience | |
| courses | a complete lesson plan for their next topic (ask which topic - one question) | |
| coaching | a session prep pack for their next call (ask who/what - one question) | |
| mix | ask: "Which would help most this week?" and offer the three above |
The active pack's own work (5.3.0). Read packs/active.json -> active_pack, then packs/{active_pack}/pack.json in this plugin, and take its domain_skills[].
- If that list is missing or empty, or already includes
community-pulse,lesson-plannerorsession-planner, the table above is this pack's first win: use it unchanged. - Otherwise, offer each installed skill in that list (at most three) beside the table's deliverable, in the same one question, naming each by its deliverable from that skill's own
description; produce the one the owner picks by following that skill's SKILL.md.
At most ONE question before producing. If the answer is vague, make a sensible choice and say so.
Produce the REAL thing, never a sample. The deliverable must be actual, usable work the owner can deliver from or post TODAY - never a "sample", "example", "demo", or "template to show what's possible". Do not label it as a sample and do not add disclaimers like "swap the topic anytime" or "tell me your real topic and I'll build the real one". If you don't know the topic, ASK the one question - asking is always better than producing a demo. The test: would the owner walk into their next class/call/community and use this exact file? If not, it's not a first win.
Step 2 - Produce
Make it genuinely good - specific to their subject and audience, not template filler. Their words from setup should visibly appear in the work. Length: enough to use today, short enough to read in 5 minutes.
Step 3 - Save + show
- Save to {campus root}/outputs/{YYYY-MM}/first-win-{slug}.md - the outputs folder INSIDE the campus folder (next to workspace.json), never a session outputs directory.
- Show the full deliverable in chat.
- Close the loop, plainly: "That's your team's first piece of work - it's
saved in your outputs folder. You can use it as-is or tell me what to change."
Step 4 - One forward pointer (not a tour)
Offer exactly one next action, chosen by delivery type - e.g. "Want me to draft the next one, or set this up to happen weekly?" If they take the weekly offer, that becomes their first scheduled task.
Log the run to the ledger your seat resolves to (.campus-os/ledger-paths.md) (skill: "campus-first-win"). If the owner renamed their chief of staff, speak as that name.
Rules
- Under 10 minutes from install to saved deliverable. That's the promise; if a
step threatens it, cut the step.
- No configuration talk during the win. Enrichment starts NEXT session.
- Never fake specificity - if you don't know their niche's jargon, write plainly
rather than guessing wrong.
File anchoring (critical)
Kernel preamble §2 applies: references/kernel-preamble.md.
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/campus-first-win/SKILL.md
Popularity
- 7-day trend score
- 5
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/trainingsites/campus-ai-os/campus-first-win)