build
S4 of see-it-ship-it: build one feature at a time, simplest stack that works, checkpoint before anything risky, rules kept apart…
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 before polish, risks and costs flagged, free alternatives first. Internal planning is written in iLang; the person never sees it.
$ npx -y skills add ilang-ai/autocode --skill plan --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/planContext preview
The summary Claude sees to decide when to auto-load this skill.
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 before polish, risks and costs flagged, free alternatives first. Internal planning is written in iLang; the person never sees it.
name: plan description: 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 before polish, risks and costs flagged, free alternatives first. Internal planning is written in iLang; the person never sees it. 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} ::PRIOR{dimension:v1_scope|default:core_tier_only_one_session|authority:system|scope:@MVP}
::GENE{plan|conf:confirmed|scope:global} T:steps=5-15|each=2-5min|dependency_order_first T:each_step_ends_in_something_clickable_or_testable T:estimate_in_minutes|total_at_end|buffer=+20% T:core_function_first|revenue_before_aesthetics|mvp_then_iterate T:flag_third_party_dependencies_rate_limits_and_cost_traps|free_alternatives_first T:large_task⇒phases_with_milestones_and_a_shipping_date T:compare_with_a_mid_level_programmer_only_at_milestones|time_and_money|realistic T:internal_plans_in_ilang|user_never_sees_compressed_form T:mvp_from_product_study_is_the_plan|when:S2_ran A:step_over_5_minutes⇒split_further A:monolithic_spec⇒reject A:say_it_depends⇒give_a_concrete_number A:all_features_at_once⇒sequential A:polish_before_core⇒reject A:roadmap_without_a_shipping_date⇒add_one A:exaggerate_the_comparison⇒realistic A:filler_words_in_internal_notes⇒remove A:meaning_loss_in_compression⇒reject
::ACTIVATE{plan} ON:task_scope>30min ON:plan_created ON:multiple_features_requested ON:external_api_detected ON:S2-study.gate_passed
::STATE{@STEPS, ordered steps, each with minutes and what the person will see}
[PLAN:@STEPS|whr=dependency_order|lim=15]=>[CHEK:@STEPS|whr=each_clickable]=>[Ω:@USER|fmt=text|lim=5]
::EVIDENCE{id:plan1|deliverable:steps|kind:text|ref:the_plan_shown|verified_by:@USER|result:they_said_go}
::SAY{@AUTOCODE→@USER}{ output: "1. 搭框架(2分钟)2. 注册页面(5分钟)3. 登录功能(5分钟)4. 上线(5分钟)… 大概 20 分钟搞定。" risk: "这个功能要用 X 的接口,免费额度每天 100 次,够我们用;超了我会先说。" }
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.
S4 of see-it-ship-it: build one feature at a time, simplest stack that works, checkpoint before anything risky, rules kept apart…
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,…
Study an existing product from its public pages and turn it into an evidence-backed blueprint plus an MVP cut that AutoCode can…