docs(getting-started): end-to-end verify all tutorial content; add per-domain skills catalog#3375
Merged
Merged
Conversation
…n skills catalog Every load-bearing claim on build-with-claude-code and your-first-project was re-verified in a freshly scaffolded 16.0.0-rc.1 project: the three tutorial example files pass os validate verbatim, the bare-reference error text matches the real gate output word-for-word, the curl flow (sign-in → create → select/sort/top query) runs as written, and the MCP endpoint/tool table matches the registered tool surface. build-with-claude-code: - prerequisites no longer claim pnpm is required (npm suffices; matches your-first-project); stale 'Create ObjectStack v6.x' banner replaced with the current scaffolder output incl. Next steps - document that the agent wires new files into objectstack.config.ts (actions:/views: arrays + objects barrel) — previously unmentioned, and the first thing to check when authored metadata doesn't show up - new section 'Beyond the first app — one skill per metadata domain': maps all nine skills to the metadata domain they drive with a copy-paste example prompt each, plus the two habits that keep the loop reliable (validate gates every domain; re-run skills add after spec upgrades) - surface the os lint --score quality rubric in the iterate step your-first-project: - surface os lint --score after the validate gate - new section '7. Hand the project to your agent' bridging the manual path to the skills-driven loop with a cross-domain example prompt and a link to the per-domain catalog Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Kj2MJEPXoUXC4LiC3XYJc
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
Every command/subcommand/flag on the page is now verified against packages/cli/src/commands/; validate, info, compile, build, lint --score and --help were executed in a real 16.0.0-rc.1 consumer app and the printed samples updated to match actual output. - os start: the page claimed auth is silently skipped without a secret — reality is --auth-secret > OS_AUTH_SECRET > an auto-generated secret persisted at <home>/auth-secret, so auth works out of the box (start.ts readOrCreateAuthSecret); documented the full artifact resolution chain incl. config auto-compile and the empty-kernel fallback, and the missing --home flag. - Replaced the monorepo-internal '../objectui Vite server' Console note with the consumer reality (prebuilt bundle served at /_console/). - os compile / os info samples regenerated from real runs; os dev gains --fresh/--seed-admin rows + the persistent dev-DB note and the seeded admin / MCP endpoint mention; added the missing os build alias and an os lint section (--score/--fix/--json); serve auth tier note gains the dev-fallback-secret parenthetical. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Kj2MJEPXoUXC4LiC3XYJc
All ten patterns were dropped verbatim into a scaffolded 16.0.0-rc.1 app
and run through npm run validate — twice: as-published (passes, proving
the gate does NOT catch these) and as-fixed. The fixes target
runtime-fatal issues invisible to validate:
- Both flow patterns lacked the start-node trigger binding
(config.objectName + triggerType: 'record-after-update' …) — they
validated but never fired; approval-flow nodes (auto_approve /
request_approval / update_record) had no config at all, making them
no-ops. Bindings, node configs, status: 'active', and a CEL start
condition added.
- reference: 'user' → 'sys_user' (no 'user' object exists; lookups
silently never resolve).
- notify recipient '{record.assigned_to.email}' → '{record.assigned_to}'
(recipients are user ids; path-walking a string yields empty
recipients and a node error).
- Formula field: bare 'quantity * unit_price' →
'record.quantity * record.unit_price' + returnType (bare identifiers
resolve to null in the CEL scope).
- Cursor-pagination section replaced with keyset pagination (cursor is
schema-reserved, zero engine/driver execution — same finding as
PR #3374).
- Frontmatter description no longer promises realtime; import callout
gains definePermissionSet.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012Kj2MJEPXoUXC4LiC3XYJc
…eference quick-start (Anatomy) verified claim-by-claim against the Zod sources and the blank template — no changes needed. validating-metadata: - added a real-shaped clean-run output sample (step lines match validate.ts printStep calls verbatim); - corrected gate-parity wording: compile/build shares the core validators, while os validate additionally runs list-view modes (ADR-0053), view-container shape, and JSX/React page-source checks (ADR-0080/0081); - removed the monorepo-internal pnpm --filter instruction. quick-reference (~45 corrections, each grep-verified in packages/spec/src): - removed rows for nonexistent compliance.zod.ts / masking.zod.ts and a duplicated Service Registry row; moved Dataset from Data to UI; - expanded the AI table 3→11 rows (Skill, Tool, MCP, Embedding, Knowledge Source/Document, Usage, Solution Blueprint — the section advertised RAG/cost tracking with zero matching rows); - added Environment (cloud) and Expression (shared) rows; - fixed ~35 stale Key Schemas names to real exports (QueryOptions→ QueryAST, PluginHook→PluginLifecycle, Auth→LoginRequest/Session, Channel→RealtimeEvent, Package→MarketplaceListing/PackageSubmission, all seven kernel plugin-* rows, …) and updated table counts. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Kj2MJEPXoUXC4LiC3XYJc
os-zhuang
marked this pull request as ready for review
July 21, 2026 02:13
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
概述
#3374(skills 全面审核)的后续:按同样的“对照真实运行时逐字核实”标准优化整个入门(getting-started)板块,让开发者能顺着 skills 流程完成各类元数据应用的开发。覆盖 7 个内容页(其余 4 页经陈旧特征清单扫描 + 逐项核实确认无需改动)。
验证方法
所有示例代码放进全新
create-objectstack@16.0.0-rc.1脚手架项目实际运行;所有 CLI 命令对照packages/cli/src/commands/源码逐条核实并实跑采样;所有速查表行 grep 核实到 Zod 源文件。各页修改
build-with-claude-code + your-first-project(第一批)
os validate逐字通过;curl 流程原样可运行;MCP 工具表与注册面一致objectstack.config.ts的说明os lint --scorecli.mdx
validate/info/compile/build/lint --score/--help实跑并对齐输出样例os start认证描述(原文称无 secret 则跳过认证;实际自动生成并持久化 secret,开箱即用);补全 artifact 解析链、--home、--fresh/--seed-admin;删除 monorepo 内部的../objectuiVite 说明common-patterns.mdx
reference: 'user'→sys_user;notify 收件人是用户 id 而非.email路径;公式字段裸标识符 →record.前缀;cursor 分页 → keysetvalidating-metadata.mdx
quick-reference.mdx(约 45 处)
compliance.zod.ts/masking.zod.ts行;Dataset 移到 UI 表;AI 表 3→11 行补齐 RAG/成本跟踪相关 schema;修正约 35 个过时的 Key Schemas 名称(每个替换名都 grep 核实)quick-start.mdx
验证
pnpm --filter docs build本地通过(两轮)🤖 Generated with Claude Code
https://claude.ai/code/session_012Kj2MJEPXoUXC4LiC3XYJc