fix(ci): lint-readme green — lychee.toml parse error, markdownlint config + content fixes

.github/workflows/pr-check.yml

@@ -20,7 +20,9 @@ jobs:
         with:
           node-version: 20
       - run: npm install
-      - run: npx playwright install --with-deps chromium
+      # playwright.config.js runs chromium, firefox and webkit; installing only
+      # chromium made every firefox/webkit test fail at launch on every PR.
+      - run: npx playwright install --with-deps chromium firefox webkit
 
       - name: HTML validation
         run: |

.markdownlint.yml

@@ -0,0 +1,19 @@
+# markdownlint configuration, ported from embeddedos-org/embeddedos-org
+# (9475a30) together with the lint-readme workflow it belongs to (6fc5b85).
+# markdownlint-cli auto-discovers this file, so the workflow needs no flags.
+
+# MD013 (line-length): prose stays at the 80-column guideline; tables are
+# exempt because index/reference rows are inherently long.
+MD013:
+  line_length: 80
+  tables: false
+
+# MD060 (table-column-style): any one pipe style per table, consistently.
+MD060:
+  style: consistent
+
+# MD024 (no-duplicate-heading): the ecosystem guide repeats the same
+# subsection set (Overview, Key Features, Architecture, ...) under every
+# chapter. Duplicates are still reported among siblings of one parent.
+MD024:
+  siblings_only: true

CHANGELOG.md

@@ -8,7 +8,8 @@ This is the synchronized production release across all 18 EmbeddedOS-org repos.
 
 - Refreshed governance: LICENSE, NOTICE, CITATION.cff, SECURITY.md
 - CI/CD pipelines hardened: release.yml, book-build.yml, video-build.yml, deploy-pages.yml
-- Release artifacts produced for: Linux x64/arm64, macOS x64/arm64, Windows x64, Docker, plus per-repo embedded/mobile/extension targets
+- Release artifacts produced for: Linux x64/arm64, macOS x64/arm64, Windows x64,
+  Docker, plus per-repo embedded/mobile/extension targets
 - mdBook documentation built and deployed to GitHub Pages
 - Promo video rendered and attached as a release asset
 
@@ -20,7 +21,8 @@ This is the synchronized production release across all 18 EmbeddedOS-org repos.
 
 - Refreshed governance: LICENSE, NOTICE, CITATION.cff, SECURITY.md
 - CI/CD pipelines hardened: release.yml, book-build.yml, video-build.yml, deploy-pages.yml
-- Release artifacts produced for: Linux x64/arm64, macOS x64/arm64, Windows x64, Docker, plus per-repo embedded/mobile/extension targets
+- Release artifacts produced for: Linux x64/arm64, macOS x64/arm64, Windows x64,
+  Docker, plus per-repo embedded/mobile/extension targets
 - mdBook documentation built and deployed to GitHub Pages
 - Promo video rendered and attached as a release asset
 
@@ -32,6 +34,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 ## [0.1.0] - 2026-03-31
 
 ### Added
+
 - Initial release of embeddedos-org.github.io
 - Complete CI/CD pipeline with nightly, weekly, and QEMU sanity runs
 - Full cross-platform support (Linux, Windows, macOS)

CLAUDE.md

@@ -96,7 +96,7 @@ Format and worked examples: [VERIFY.md](./VERIFY.md).
 
 ---
 
-# Repository
+## Repository
 
 EoS Website
 
@@ -132,7 +132,6 @@ When two of these documents conflict, the earlier wins:
 Say that a conflict exists rather than resolving it silently. A rule two people
 read differently is a defect in the rule, and it will recur until it is fixed.
 
-
 ## Node
 
 Node.js is not installed system-wide in this WSL environment. Put the local

CODE_OF_CONDUCT.md

@@ -1,4 +1,5 @@
 # Code of Conduct — Contributor Covenant v2.1
+
 We are committed to providing a welcoming and inclusive environment.
-See: https://www.contributor-covenant.org/version/2/1/code_of_conduct.html
-Report violations to: conduct@embeddedos.org
+See: <https://www.contributor-covenant.org/version/2/1/code_of_conduct.html>
+Report violations to: <conduct@embeddedos.org>

CONTRIBUTING.md

@@ -24,4 +24,5 @@ Follow [Conventional Commits](https://www.conventionalcommits.org/).
 
 ## License
 
-By contributing, you agree that your contributions will be licensed under the MIT License.
+By contributing, you agree that your contributions will be licensed under the
+MIT License.

MEMORY.md

@@ -1,8 +1,8 @@
 <!-- generated: eos-ai-scaffold -->
 # Memory
 
-Durable context for `embeddedos-org.github.io` — decisions and constraints that outlive one
-session and are not recoverable from the code or the git history.
+Durable context for `embeddedos-org.github.io` — decisions and constraints that
+outlive one session and are not recoverable from the code or the git history.
 
 Write here when a future reader would otherwise repeat an argument that was
 already settled, or repeat a mistake that was already made.
@@ -34,10 +34,11 @@ notices the obvious-looking alternative.
 | —    | None recorded yet. | — | — |
 
 <!-- Example of the level of detail worth recording:
-| 2026-03-14 | Queue writes in-process rather than via Redis | Deploy target has no
-network sidecar; measured throughput was sufficient at 4x expected peak |
-Redis Streams — rejected on operational cost, not on capability. Revisit if
-peak exceeds 8x. |
+     2026-03-14 — Queue writes in-process rather than via Redis.
+     Reason: the deploy target has no network sidecar, and in-process
+     throughput measured fine at 4x expected peak.
+     Rejected alternative: Redis Streams (operational cost, not capability).
+     Revisit if peak exceeds 8x.
 -->
 
 ## Constraints

README.md

@@ -26,7 +26,7 @@ CSS, and vanilla JavaScript, no build step required. Package name:
 
 Supporting content:
 
-```
+```text
 js/          animations.js, ebot-chat.js, search.js, site-chrome.js
 stacks/      Technology stack pages (eai-edge.html, index.html)
 eApps/       eApps index + icons
@@ -55,7 +55,7 @@ Serve the site locally (static file server on port 8080):
 npm run serve      # npx http-server . -p 8080 -s
 ```
 
-Then open http://localhost:8080.
+Then open <http://localhost:8080>.
 
 ## Test
 

SECURITY.md

@@ -8,7 +8,7 @@
 
 ## Reporting a Vulnerability
 
-**Email:** security@embeddedos.org
+**Email:** <security@embeddedos.org>
 
 > Do **NOT** open public GitHub issues for security vulnerabilities.
 
@@ -29,4 +29,6 @@
 
 ## Safe Harbor
 
-We consider security research conducted in good faith to be authorized and will not pursue legal action against researchers who follow responsible disclosure practices.
+We consider security research conducted in good faith to be authorized and will
+not pursue legal action against researchers who follow responsible disclosure
+practices.

TASKS.md

@@ -1,9 +1,10 @@
 <!-- generated: eos-ai-scaffold -->
 # Tasks
 
-Working ledger for `embeddedos-org.github.io`. The planner writes entries; each owning role
-updates its own row. Roles are in [AGENTS.md](./AGENTS.md), the workflow in
-[ORCHESTRATION.md](./ORCHESTRATION.md), the gate in [VERIFY.md](./VERIFY.md).
+Working ledger for `embeddedos-org.github.io`. The planner writes entries; each
+owning role updates its own row. Roles are in [AGENTS.md](./AGENTS.md), the
+workflow in [ORCHESTRATION.md](./ORCHESTRATION.md), the gate in
+[VERIFY.md](./VERIFY.md).
 
 Status is one of: `todo`, `in-progress`, `blocked`, `review`, `done`.
 
@@ -55,7 +56,8 @@ Verification
 
 ## Verification commands for this repository
 
-These commands were derived from the manifests at the repository root. Confirm one works before relying on it; a listed script may still be a stub.
+These commands were derived from the manifests at the repository root. Confirm
+one works before relying on it; a listed script may still be a stub.
 
 | Check | Command | Default state |
 |-------|---------|---------------|

VERIFY.md

@@ -8,7 +8,8 @@ Do not report success without evidence. This file defines what counts.
 - `PASS` — the check ran and succeeded; the output was seen.
 - `FAIL` — the check ran and failed; quote the actual output.
 - `NOT RUN` — the check was not executed.
-- `UNKNOWN` — the check cannot be run in this environment, or its result cannot be interpreted.
+- `UNKNOWN` — the check cannot be run in this environment, or its result cannot
+  be interpreted.
 
 `NOT RUN` and `UNKNOWN` are acceptable answers, and reporting one honestly is
 never a failure. A guess dressed as `PASS` is.
@@ -37,26 +38,26 @@ Anything short of that is `NOT RUN` or `UNKNOWN`.
 
 > `pnpm test:unit` — exit 0. 142 passed, 0 failed, 3 skipped. The 3 skips are
 > pre-existing and unrelated to this change. → `PASS`
-
+>
 > `pnpm check` — exit 2. `src/api/user.ts(41,7): error TS2345: Argument of type
 > 'string | undefined' is not assignable to parameter of type 'string'.`
 > → `FAIL`
-
+>
 > No linter is configured in this repository. → `NOT RUN`
-
+>
 > The e2e suite needs a running database, which is not available here.
 > → `UNKNOWN` — e2e behaviour is unverified.
 
 ### Not valid
 
 > The tests should pass now.
-
+>
 > I've made the change, so the build will work.
-
+>
 > Everything looks good.
-
+>
 > `PASS` (no command, no output)
-
+>
 > Tests pass — *(when the suite was never run, only written)*
 
 The difference is not tone. The first set can be checked by someone else; the