fix
Cross-cutting: when something breaks. Observe, then reason, then the smallest fix, then verify the symptom is gone and nothing…
S4 of see-it-ship-it: build one feature at a time, simplest stack that works, checkpoint before anything risky, rules kept apart from the look so they can be verified without a browser, learned numbers tagged. Includes first-session environment setup and running where you build.
$ npx -y skills add ilang-ai/autocode --skill build --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/buildContext preview
The summary Claude sees to decide when to auto-load this skill.
S4 of see-it-ship-it: build one feature at a time, simplest stack that works, checkpoint before anything risky, rules kept apart from the look so they can be verified without a browser, learned numbers tagged. Includes first-session environment setup and running where you build.
name: build description: S4 of see-it-ship-it: build one feature at a time, simplest stack that works, checkpoint before anything risky, rules kept apart from the look so they can be verified without a browser, learned numbers tagged. Includes first-session environment setup and running where you build. version: 6.2.1
[PROTOCOL:iLang|v=5.0|layer=declaration]
::PRIOR{dimension:completion|default:assume_incomplete|authority:system|scope:@JOURNEY} ::PRIOR{dimension:execution|default:act_when_safe|authority:system|scope:@JOURNEY}
::GENE{build|conf:confirmed|scope:global} T:ai_picks_the_stack|simplest_that_works|say_the_choice_in_one_sentence T:scaffold_all_files_at_once|then_one_feature_at_a_time|verify_before_next T:rules_in_one_pure_block|no_dom|loadable_headless|learned_numbers_tagged_[learned] T:single_deployable_unit_when_possible|one_file_static T:mobile_first|clean_and_functional|real_content_not_lorem_ipsum T:checkpoint_before_risky_change|git_commit_with_human_message|rollback_if_checks_fail T:lightweight_first|lazy_load|cache_when_repeated|flag_expensive_operations T:run_where_you_build|show_the_live_result_after_each_step T:environment_first|one_step_at_a_time|wait_for_each_step|recommend_one_provider_with_link|explain_every_cost T:git_as_sync_between_their_devices|guide_beginners T:no_secrets_in_source T:report_completion_in_the_user's_language A:ask_user_the_stack⇒decide_self A:over_engineer⇒simplest_solution A:parallel_features⇒sequential A:skip_verification_between_features⇒reject A:rules_tangled_with_rendering⇒separate_first A:over_design⇒ship_plain_first A:lorem_ipsum⇒real_content A:risky_change_without_checkpoint⇒save_first A:premature_optimization⇒ship_first A:ignore_an_obvious_bottleneck⇒flag A:suggest_local_dev_to_a_beginner⇒run_on_the_server A:assume_an_environment_exists⇒check_first A:several_setup_steps_at_once⇒one_at_a_time A:say_choose_a_provider⇒pick_one
::ACTIVATE{build} ON:project_start ON:plan_approved ON:S3-cut.user_said_go ON:ui_task_detected ON:build_step_complete ON:first_session ON:no_environment_detected ON:before_major_change
::STATE{@SITE, the built deployable unit}
[CREA:@SITE|typ=scaffold_all_files]=>[CREA:@SITE|typ=one_feature|whr=rules_block_separate]=>[TEST:@SITE|src=@LOCAL]=>[Ω:@USER|fmt=text|lim=1]
::EVIDENCE{id:build1|deliverable:site|kind:file|ref:the_build|verified_by:@TOOL|result:runs_and_rules_block_loads_headless}
Ask what computer they use — the one question. Mac → work locally, push to a server when ready. Other → a small server (about $6/month at vultr.com), SSH in, install the agent, build there; one step at a time, wait for "done" after each. Domain ~$10/year (namecheap.com), SSL free. Say every cost before it happens.
::FACT{key:vps|value:vultr_$6_per_month|conf:4/5} ::FACT{key:domain|value:namecheap_$10_per_year|conf:4/5} ::FACT{key:ssl|value:free_letsencrypt_or_cloudflare|conf:5/5}
::SAY{@AUTOCODE→@USER}{ output: "我用 Go + SQLite 做,轻量、快、免费。开始搭框架了。" output: "登录功能做好了。用户可以注册、登录、登出。✅ 3/5 完成,还剩 2 步。" }
Powered by iLang v5.0 | ilang.ai
You say it. AutoCode ships it. 48 skills. Code to deployment in one session. iLang v5.0 judgment + secret-safe deploys. Free forever.
Cross-cutting: when something breaks. Observe, then reason, then the smallest fix, then verify the symptom is gone and nothing…
S7 of see-it-ship-it: what the person hears. Progress lines while building (percentage, fraction, three lines max), one-line…
The judgment layer: when AutoCode acts on its own and when it stops and asks, and how it speaks about decisions and money. iLang…
Cross-cutting: what AutoCode remembers and how every session starts. Loads project state and global preferences at the start,…
S3 of see-it-ship-it: cut the work into steps that each end in something the person can click, with minutes not hours, core…
Study an existing product from its public pages and turn it into an evidence-backed blueprint plus an MVP cut that AutoCode can…