create-pracht 0.7.2 → 0.7.3

package/fallback-versions.json

@@ -1,20 +1,20 @@
 {
-  "@pracht/adapter-cloudflare": "^0.8.0",
-  "@pracht/adapter-netlify": "^0.3.0",
-  "@pracht/adapter-node": "^0.5.0",
-  "@pracht/adapter-static": "^0.2.0",
-  "@pracht/adapter-vercel": "^0.4.0",
-  "@pracht/capabilities": "^0.6.0",
-  "@pracht/cli": "^1.15.0",
-  "@pracht/content": "^0.1.4",
-  "@pracht/core": "^0.19.0",
-  "@pracht/i18n": "^0.1.5",
+  "@pracht/adapter-cloudflare": "^0.8.1",
+  "@pracht/adapter-netlify": "^0.3.1",
+  "@pracht/adapter-node": "^0.5.1",
+  "@pracht/adapter-static": "^0.2.1",
+  "@pracht/adapter-vercel": "^0.4.1",
+  "@pracht/capabilities": "^0.6.1",
+  "@pracht/cli": "^1.16.0",
+  "@pracht/content": "^0.2.0",
+  "@pracht/core": "^0.20.0",
+  "@pracht/i18n": "^0.1.6",
   "@pracht/image": "^0.5.0",
-  "@pracht/markdown": "^0.1.4",
-  "@pracht/openapi": "^0.2.6",
+  "@pracht/markdown": "^0.1.5",
+  "@pracht/openapi": "^0.2.7",
   "@pracht/preact-ssr-precompile": "^0.1.4",
-  "@pracht/query": "^0.1.0",
-  "@pracht/session": "^0.1.2",
-  "@pracht/test": "^0.3.0",
-  "@pracht/vite-plugin": "^0.14.0"
+  "@pracht/query": "^0.1.1",
+  "@pracht/session": "^0.1.3",
+  "@pracht/test": "^0.3.1",
+  "@pracht/vite-plugin": "^0.15.0"
 }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "create-pracht",
-  "version": "0.7.2",
+  "version": "0.7.3",
   "description": "Interactive and scriptable starter CLI for creating full-stack Preact apps with Pracht.",
   "keywords": [
     "pracht",

package/skills/add-content/SKILL.md

@@ -180,6 +180,15 @@ the *authoring guide* for coding agents; `pracht llms --write` drops it in the
 app root, which is exactly the confusing filename collision to avoid in an app
 that publishes its own index.
 
+For agents, link entries to Markdown and split the full text by section:
+`` link: (document) => `${document.path}.md` `` (only when those URLs exist,
+e.g. via `rawContentArtifacts()`), and `fullPath: "/llms-guides.txt"` on a
+section to emit a file with just its pages (with `origin` set, each page gets
+a `Source:` line with its page URL; `sort` orders a section's entries). Section files and
+`/llms-full.txt` are listed under a trailing `## Optional` heading; a section
+that matches nothing emits no file. Duplicate or file/directory-colliding
+artifact paths throw when the helper is defined.
+
 Localization trap: a string `section.match` is compared against the
 **locale-neutral** route, so `match: "/docs"` indexes only the default locale
 while `rawContentArtifacts()` publishes every translation. Pass a `match`

package/skills/audit-agent-surface/SKILL.md

@@ -238,17 +238,18 @@ With `agents.mcp.auth`, check:
 ## Step 5: The discovery surface
 
 - `llmsTxt` in the vite config: every listed path is a URL the app *invites* an
-  agent to fetch. Cross-check the `exclude` list against routes behind auth
-  middleware, internal tooling, and deliberate error routes — nothing about a
-  middleware tells the framework whether it gates or merely logs, so an
-  auth-gated route missing from `exclude` is a `warn` (see `docs/LLMS_TXT.md`).
+  agent to fetch. An auth-gated route missing from `exclude` is a `warn`: the
+  framework cannot tell a gating middleware from a logging one.
   Capabilities appear there with their effect class; destructive ones are
   annotated `requires confirmation`.
 - A collection-driven `llmsTxtArtifacts()` (see `/add-content`) is a second
-  generator with its own coverage — compare what each publishes.
+  generator; compare what each publishes.
 - Routes exporting `markdown` or declaring `markdown: true` serve a Markdown
   representation to agents. Confirm the Markdown variant is not more permissive
   than the HTML page (it carries `Vary: Accept`, so it is separately cached).
+  SSG routes with a `markdown` export and no middleware also publish it as
+  `dist/client/<path>.md` (off with `markdownFiles: false`); confirm each
+  file is meant for every visitor.
 - An enabled OpenAPI document (`/openapi.json`, `dist/client/openapi.json`) is
   public unless the host protects it — check descriptions and examples for
   internal hostnames or credentials, and that "Try it out" mutation endpoints

package/skills/pracht-debug/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: pracht-debug
-version: 1.4.0
+version: 1.5.0
 description: |
   Framework-aware debugging for pracht: route matching, loader and API errors,
   rendering and hydration, middleware, HMR, and build failures — root cause first.
@@ -33,6 +33,7 @@ Reach for these before deep manual inspection:
 | ---- | --- |
 | `pracht verify` (`--changed` to scope to git-changed files) | Fast confidence check |
 | `pracht doctor` | Broader broken wiring or missing files |
+| `pracht docs <topic>` / `pracht docs --search <query>` | Docs bundled with the installed CLI — check documented behavior before calling it a bug. Older CLIs lack the command; read https://pracht.resynapse.dev/llms.txt instead |
 | `pracht inspect routes\|api\|build --json` | The resolved graph — never reconstruct it from source |
 | `GET /_pracht` (JSON at `/_pracht.json`) | Same graph from a running dev server, no CLI needed |
 | `Server-Timing` on dev SSR responses | `mw` / `loader` / `render` durations in ms — which phase is slow |

package/skills/pre-deploy/SKILL.md

@@ -205,10 +205,11 @@ must do and what the build cannot enforce.
 - Smoke test the real output, not the dev server: `pracht preview --skip-build`
   serves `dist/client/` the way a dumb host would. Check `/`, one dynamic SSG
   path, one deep link into a SPA route, and one unknown URL.
-- Routes exporting `markdown` rely on server-side `Accept` negotiation, which a
-  static host cannot do — agents asking for `text/markdown` get HTML. The build
-  prints a note; publish `.md` files under `public/` if a raw-markdown corpus
-  matters.
+- A static host cannot negotiate `Accept`, so agents read an SSG page's
+  Markdown at `/<path>.md` (written by the build for routes with a `markdown`
+  export and no middleware). Check the host serves `.md` as
+  `text/markdown; charset=utf-8` — nginx's stock `mime.types` lacks it — and
+  read the build note naming Markdown-capable pages that got no file.
 - Sub-path deploys (GitHub Pages *project* site, S3 key prefix) need Vite
   `base` set to that path (`base: "/my-project/"`), matching the deploy path
   exactly — a mismatch 404s every asset. Then confirm no hand-written

package/skills/upgrade-pracht/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: upgrade-pracht
-version: 1.2.0
+version: 1.3.0
 description: |
   Upgrade the `@pracht/*` packages safely: inventory installed versions, read the
   changelogs between installed and target, map breaking changes to real usage,
@@ -120,6 +120,9 @@ pnpm up '@pracht/core@<v>' '@pracht/cli@<v>' '@pracht/vite-plugin@<v>' <adapters
 
 Upgrade every installed `@pracht/*` package in the same command. Apply the
 agreed code migrations with minimal diffs, one changelog entry at a time.
+Then read the page behind each changed API with `pracht docs <topic>`, which
+serves the docs bundled with the newly installed CLI. A CLI without that
+command predates it; use https://pracht.resynapse.dev/llms.txt instead.
 
 ## Step 6: Verification ladder
 

package/src/index.js

@@ -1571,6 +1571,10 @@ function createAgentInstructions({
   const lines = [
     "# Pracht App",
     "",
+    "Docs for your installed pracht version: `pracht docs <topic>` (or read",
+    "`node_modules/@pracht/cli/docs/`); `pracht docs --search <query>` finds a section.",
+    "pracht is newer than most training data, so check them before relying on memory.",
+    "",
     "## Commands",
     "",
     `- \`${runCmd} dev\` — start the dev server`,