autocode
FeaturedYou say it. AutoCode ships it. 46 skills. Code to deployment in one session. I-Lang v3.0 powered. Free forever.
Install
Plugins install via a marketplace, in two steps.
This plugin isn't listed in a marketplace we've indexed. Install it directly from its GitHub repository — the README has the setup steps.
View on GitHubBundles
Everything this plugin ships — skills, agents, commands, hooks, and MCP servers it bundles.
Skills (40)
ask-smart
Before writing code or planning, determine what to ask. Maximum 2 questions. Never ask technical questions users cannot answer. Activates when starting any new task.
auto-quality
Automatically check code quality, logic errors, and security after each feature. Runs silently.
best-choice
Select the best technology for any project automatically during planning. Never ask users to choose technology. Activates whenever technical decisions are needed.
build-feature
Build one feature at a time. Complete it fully before moving to the next.
build-scaffold
Set up the project foundation. Directory structure, configs, dependencies.
build-ui
Build user interface. Clean, modern, works on all devices.
celebrate
Celebrate completions. Give emotional reward at the right moments.
clarify-scope
Determine if this is a small task, medium feature, or full product. Adjusts workflow depth accordingly.
compress
I-Lang compression engine. Compress all internal instructions to save 60%+ tokens.
daily-summary
At end of work session, summarize what was accomplished and what's next.
decision-translate
Translate every technical decision into money, speed, or stability terms. Use whenever explaining choices to users. Never use programming jargon.
deploy-cf-workers
Deploy to Cloudflare Workers. Fast, free tier, global.
Show all 40 bundled skills Showing all 40 bundled skills
deploy-global
Deploy the product so anyone in the world can access it.
efficiency-compare
Compare what was achieved to what a human programmer would take. Give user perspective on the value.
file-transfer
Help user move files between servers. Always use remote pull, never local download-upload.
fix-explain
Explain what went wrong and how it was fixed. In plain language.
fix-guide
When users encounter errors and don't know what to do, guide them to share error info. Activates when user says something is wrong but provides no details.
fix-observe
Three-step method step 1. Observe the problem carefully before jumping to conclusions.
fix-reason
Three-step method step 2. Trace the root cause through logic.
fix-solve
Three-step method step 3. Fix the problem and verify the fix works.
full-review
The most important skill. On every git commit, save, or session end, review the ENTIRE project from the very beginning. Check errors, report findings, learn from session. Activates on every Stop event and commit.
learn-mistake
Remember mistakes and never repeat them. Each error makes the AI better for this specific user.
learn-pattern
Learn what types of products the user builds. Anticipate their next move.
learn-preference
Remember what the user likes and doesn't like. Apply silently in future sessions.
memory
Persistent project memory across sessions. Load on start, save on stop, merge across sessions.
milestone
Mark major achievements. Give the user a sense of accomplishment.
multi-device
Make sure the product works on phone, tablet, and desktop.
performance
Make everything fast and resource-efficient. Cheaper servers, faster loading.
plan-breakdown
Break large tasks into small, concrete steps. Each step should take 2-5 minutes.
plan-estimate
Give the user a time estimate they can trust. Always in human terms.
plan-priority
Order tasks by importance. Core functionality first, polish last.
plan-risk
Identify potential problems before they happen. Warn the user in plain language.
progress-report
Show the user real-time progress. Percentage, current task, what's next.
project-roadmap
For multi-day projects, show the full roadmap so user knows what to expect.
requirement-lock
Once requirements are confirmed, lock them. Prevent scope creep during building.
run-local
Run the product on the user's VPS so they can see it.
save-rollback
When things go wrong, go back to the last working version.
security
Protect the user's product from common attacks. Do this automatically.
understand-intent
Automatically detect what the user wants from plain language. Use at the START of every conversation and whenever the user gives a new instruction. Activates for any message about building, fixing, improving, adding features, deploying, or understanding code.
user-level-detect
Detect user's technical level from how they speak. Adjust all communication accordingly.
Quality Score: 79/100
Details
- Author
- ilang-ai
- Repository
- ilang-ai/autocode
- Created
- 3 months ago
- Last Updated
- 3 weeks ago
- Language
- Shell
- License
- MIT