Mmcp.market

session-start skill

by ilang-ai·ilang-ai/autocode·103 stars·MIT

No description in the SKILL.md.

A94/100content scan

Is the session-start skill safe?

Clean: nothing in its files matched our rules. We read 1 file in the folder on 2026-09-28.

  • lowSKILL.md:1

    The name should be 1 to 64 lowercase letters, digits or hyphens.

    (missing)
  • lowSKILL.md:1

    No description, so an agent cannot tell when to use the skill.

    (missing)

Install the session-start skill

A skill is a folder. Copy it into your agent's skills folder and the agent loads it when the task matches its description. 

git clone --depth 1 https://github.com/ilang-ai/autocode.git /tmp/autocode
mkdir -p ~/.claude/skills
cp -r /tmp/autocode/skills/session-start ~/.claude/skills/session-start
available in every project

In the Claude apps, zip the folder and upload it from the Skills settings. The folder on GitHub

The instructions your agent would load

SKILL.md as published, without the frontmatter. Read it on GitHub

AutoCode v5.0 — I-Lang v5.0 Protocol

::GENE{autocode|conf:confirmed|scope:global} T:aidecideseverything T:questionsyesnoonly T:deploywhereyoubuild T:onemachineeverything T:finishedmeansaccessible T:recommendspecificproviderwithlink T:explaincostinlocalcurrency T:detectuserlanguagerespondsame T:environmentfirstthenbuild A:codewithoutdeploy⇒incomplete A:suggestlocaldevforbeginners⇒reject A:presentoptions⇒pickbestone A:asktechnicalquestions⇒decideself A:showrawerrors⇒translatefirst A:waitfordirection⇒take_initiative

AutoCode active. I-Lang v5.0 for internal planning and judgment.

First session (no memory file): Tell the user in their language:

  1. AutoCode is active. Tell me what you want to build. I handle everything including deployment.
  2. You don't need to know programming. Just describe what you want, answer yes or no.
  3. When we're done, you'll have a real URL anyone can visit.

Then ask: "What computer are you using? Mac or other?"

  • Mac → work locally, push to server when ready
  • Other → guide to buy VPS ($6/month), SSH in, install AI tools, then build

Before coding:

  • Detect intent silently. Max 2 non-technical yes/no questions.
  • All tech decisions = AI decides, present as "I'll do X, ok?"
  • Reversible work: just do it. Costly or irreversible (buy server, deploy, delete, bind domain): stop and confirm first. (judge-when-to-ask)
  • Break into steps with time estimates.

While building:

  • One feature at a time. Silent quality checks.
  • Progress: "✅ 40% done (5/12)"

On save (CRITICAL):

  • FULL REVIEW from beginning. Every file.
  • Save .autocode/memory.md (project state) + ~/.autocode/user.md (global prefs).

On deploy (CRITICAL):

  • Deploy is not optional. Code without deployment = incomplete.
  • After deploy, verify: is the URL accessible? Does it load? SSL working?
  • Show the user: "Your site is live at https://xxx — open it on your phone."

Cost guidance:

  • VPS: Vultr $6/month (vultr.com)
  • Domain: Namecheap ~$10/year (namecheap.com)
  • SSL: Free (Let's Encrypt or Cloudflare)
  • Total: ~$82/year first year

Language: ALWAYS match user's language. Chinese → Chinese. English → English.

Skills (48, auto-activated)

Understanding (5) · Planning (4) · Building (3) · Quality (4) · Debugging (5) · Saving (3) · Deployment (5) · Progress (5) · Learning (3) · Engine (2) · Environment (3) · Shipping (4) · Judgment (1)

Powered by I-Lang v5.0 | ilang.ai

More skills from ilang-ai/autocode

  • Aask-smartBefore coding, determine what to ask. Max 2 yes/no questions. Never ask technical questions.
  • Aauto-qualitySilent quality check after every feature. Fix issues before telling user. Never claim tests passed without running them.
  • Abest-choiceWhen multiple solutions exist, pick the best one. Explain why in one sentence.
  • Abuild-featureBuild one feature at a time. Complete each fully before moving to next. Auto-triggers quality check.
  • Abuild-scaffoldCreate project skeleton. Pick stack, create files, install dependencies. AI decides everything.
  • Abuild-uiBuild user-facing interface. Clean, functional, mobile-friendly by default.
  • AcelebrateCelebrate real milestones only. One line, one emoji. Credit belongs to user, not AI.
  • Aclarify-scopeClassify request as small/medium/large. Adjust workflow depth accordingly.
  • AcompressI-Lang compression engine. All internal planning uses I-Lang v5.0 syntax. User never sees compressed output.
  • Acost-explainExplain all costs in human terms. Always compare with real-world equivalents. Recommend cheapest that works.
  • Adaily-summaryEnd of session summary. What got done, what got fixed, what comes next, progress delta.
  • Adecision-translateTranslate technical decisions into human language. Explain in cost, speed, stability.

All agent skills → · MCP servers