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.
git clone --depth 1 https://github.com/smicolon/ai-kitWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/smicolon/ai-kit/onboard)<a href="https://agentmods.dev/commands/smicolon/ai-kit/onboard"><img src="https://agentmods.dev/badge/commands/smicolon/ai-kit/onboard.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00023 | $0.02293 |
| Opus 5 | $0.00012 | $0.01146 |
| Sonnet 5 | $0.00005 | $0.00459 |
| Haiku 4.5 | $0.00002 | $0.00229 |
Grade A, and why
onboard 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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 314 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Engineer Onboarding
Run a guided onboarding flow that personalizes project guidance based on the engineer's background and their first task.
Steps
1. Parse Arguments
Extract from user input:
- --quick: Skip skill assessment, assume mid-level generalist (optional)
- --task: Pre-specify first task to skip task question (optional)
2. Introduction
Briefly explain what will happen:
Welcome! I'll get you up to speed on this project in a few minutes.
Here's how this works:
1. A few questions about your background (so I don't over-explain things you know)
2. I'll analyze the project automatically
3. You'll get personalized guidance + a cheat sheet
Let's start.
3. Engineer Assessment (skip if --quick)
Ask these core questions using AskUserQuestion:
Question 1 - Primary Stack: "What's your primary language/framework and roughly how long have you been using it?"
Question 2 - Other Skills: "What other languages or frameworks are you comfortable with?"
Question 3 - New Territory: "Looking at this project, what feels completely new or unfamiliar to you?"
Adaptive Follow-ups (only ask if gaps detected):
After analyzing the answers, ask follow-ups ONLY for detected gaps:
-
If the project uses frontend tech and engineer is backend-only: "Any experience with component-based UI frameworks (React, Vue, etc.)?"
-
If the project uses backend tech and engineer is frontend-only: "Have you worked with ORMs, REST APIs, or database design?"
-
If the project has tests and engineer didn't mention testing: "How comfortable are you with writing tests / TDD?"
-
If the project uses TypeScript and engineer only mentioned JavaScript: "How's your TypeScript experience — types, generics, strict mode?"
Maximum 2 follow-up questions. Keep the assessment fast.
4. Project Analysis (automatic)
Analyze the project silently. Do NOT ask the engineer anything here.
4a. Detect Project Type:
# Frontend
[ -f "package.json" ] && grep -q '"next"' package.json && echo "nextjs"
[ -f "package.json" ] && grep -q '"nuxt"' package.json && echo "nuxtjs"
[ -f "package.json" ] && grep -q "@tanstack/react-router" package.json && echo "tanstack-router"
[ -f "package.json" ] && grep -q '"hono"' package.json && echo "hono"
# Backend
[ -f "manage.py" ] && echo "django"
[ -f "package.json" ] && grep -q "@nestjs/core" package.json && echo "nestjs"
# Mobile
[ -f "pubspec.yaml" ] && echo "flutter"
# Auth
[ -f "package.json" ] && grep -q '"better-auth"' package.json && echo "better-auth"
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.
- 4d ago First seen · 314 lines · 23 tokens per session scan A bc44de198747
onboard is a command published in the GitHub repository smicolon/ai-kit (6 stars, last pushed 4d ago), licensed MIT. It adds 23 tokens to every session and 2,293 once invoked, about $0.0001 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 2026-09-03.
Other commands, from other repositories
adoption-plan
Plan personalizado de adopción de Savia por rol — qué aprender, en qué orden.
adoption-sandbox
Entorno seguro de práctica — experimentar con Savia sin miedo a errores.
gsd:help
Show available GSD commands and usage guide.
structural-integrity
Audit and maintain Datacore folder structure, detect issues, optionally fix.
triage-github
Scan GitHub repos for activity, generate a summary, and create actionable GTD tasks.
execute-plan
Execute implementation plans with structured review checkpoints. Use when you have a plan document and want faithful, evidence-based execution.