@@ -3,11 +3,11 @@ You are working on Developer Tools Directory, a meta-repository that catalogs, s
 ## Repo structure
 
 - `registry.json` -- single source of truth for all 9 tool repos (name, type, counts, links, status)
-- `standards/` -- 18 Markdown docs defining conventions (folder structure, CI/CD, manifests, pages, commits, README, AGENTS.md, versioning, and more)
+- `standards/` -- Markdown docs defining conventions (folder structure, CI/CD, manifests, pages, commits, README, AGENTS.md, versioning, and more)
 - `scaffold/create-tool.py` -- Python repo generator using Jinja2 templates
 - `scaffold/templates/` -- Jinja2 templates producing a standards-compliant repo
 - `site-template/` -- shared GitHub Pages build system for tool repos (build_site.py + template.html.j2 + fonts)
-- `docs/` -- static GitHub Pages catalog site (index.html, style.css, script.js)
+- `docs/` -- static GitHub Pages catalog site (single index.html with inline CSS and JS)
 - `assets/` -- logo image
 - `.github/workflows/` -- CI/CD for this repo (validate, pages, stale, codeql, dependency-review, release, release-drafter, label-sync)
 - `AGENTS.md` -- AI agent guidance for this repo
@@ -66,8 +66,7 @@ Use conventional commits. Prefix determines changelog category:
 ## When editing docs/
 
 - `docs/index.html` is the catalog site entry point.
-- `docs/script.js` fetches `registry.json` at runtime and renders tool cards. It also has an embedded fallback copy.
-- `docs/style.css` is the full stylesheet -- dark theme, responsive, card layout.
+- All CSS and JS are inline in `docs/index.html`. Cards, filters, and stats are pre-rendered by `scripts/sync_from_registry.py`; the inline script adds search and filtering.
 - No external dependencies (CDNs, frameworks). Vanilla HTML/CSS/JS only.
 - The `pages.yml` workflow copies `registry.json` and `assets/` into `docs/` at deploy time.
 
 
@@ -20,11 +20,13 @@
 #   5. Writes a step summary and exposes counts as outputs so the calling
 #      workflow can branch on them.
 #
-# Pinning rule: tool repos MUST consume this action via @v1.7 (or a SHA),
-# never @main. The drift checker's release pipeline maintains the v1.7 tag
-# pointing at the latest 1.7.x. Q9 of the design doc explains why @main is
-# forbidden for tool-repo callers (would break every tool-repo PR on
-# unrelated meta-repo changes).
+# Pinning rule: tool repos MUST consume this action via a floating
+# @vMAJOR.MINOR tag (or a SHA), never @main. The meta-repo's release.yml
+# moves vMAJOR and vMAJOR.MINOR to each release; the scaffold emits the pin
+# derived from the meta VERSION at generation time. A minor tag stops moving
+# once the next minor ships, so bump older pins (for example @v1.9) to pick
+# up checker fixes. @main is forbidden for tool-repo callers because it would
+# break every tool-repo PR on unrelated meta-repo changes.
 # =============================================================================
 
 name: 'Agent-file drift check'
 
@@ -62,3 +62,6 @@ site-audit-evidence/pr4-sc-light.png
 site-audit-evidence/sc-dark-390.png
 site-audit-evidence/sc-dark-1280.png
 site-audit-evidence/sc-light-1280.png
+
+# Personal, machine-specific assistant instructions (not project docs)
+CLAUDE.local.md
@@ -9,7 +9,7 @@ This file tells AI coding agents how the Developer Tools Directory repo works an
 This is a **meta-repository** -- it does not contain a Cursor plugin or MCP server itself. It catalogs, standardizes, and scaffolds other TMHSDigital developer tool repos. It contains:
 
 - **`registry.json`** -- single source of truth for all tool repos. README, CLAUDE.md, and the catalog site's embedded registry are generated from it by `scripts/sync_from_registry.py`.
-- **`standards/`** -- 18 Markdown docs defining conventions for folder structure, CI/CD, plugin manifests, GitHub Pages, commits, README format, AGENTS.md format, versioning, release-doc-sync, testing, skills, rules, MCP servers, security, licensing, scope, lifecycle, and writing style.
+- **`standards/`** -- Markdown docs defining conventions for folder structure, CI/CD, plugin manifests, GitHub Pages, commits, README format, AGENTS.md format, versioning, release-doc-sync, testing, skills, rules, MCP servers, security, licensing, scope, lifecycle, and writing style.
 - **`scaffold/`** -- Python repo generator (`create-tool.py`) with Jinja2 templates that produce a fully standards-compliant new tool repo.
 - **`scripts/`** -- automation utilities. `sync_from_registry.py` regenerates all derived artifacts from `registry.json`.
 - **`site-template/`** -- shared GitHub Pages build system for tool repos. `build_site.py` reads `.cursor-plugin/plugin.json`, `site.json`, `skills/`, `rules/`, and `mcp-tools.json` from a tool repo and renders `docs/index.html` via `template.html.j2`. Self-hosts Inter and JetBrains Mono fonts. Tool repos clone this directory in CI and run the build script at deploy time.
@@ -50,7 +50,7 @@ Every PR must pass these 7 checks before it can be merged:
 
 Checks:
 - `registry.json` is valid JSON with correct schema (required fields, valid types, integer counts)
-- `docs/index.html`, `docs/style.css`, `docs/script.js` exist
+- `docs/index.html` exists
 - `scaffold/create-tool.py` compiles without syntax errors
 - Scaffold dry-run test: generates a test repo and verifies key files exist
 
@@ -61,7 +61,7 @@ Checks:
 3. Uploads `docs/` as a Pages artifact
 4. Deploys to GitHub Pages via `actions/deploy-pages`
 
-### `release.yml` (runs on push to main, ignores docs/md/standards changes)
+### `release.yml` (runs on every push to main)
 
 1. `check-version` reads `VERSION` at the repo root and compares it with the latest git tag
 2. If they match, the workflow exits cleanly - no release needed
@@ -233,8 +233,8 @@ Pure documentation -- no code. Each file documents a convention derived from ana
 ## When editing docs/ (catalog site)
 
 - `docs/index.html` is the single-page catalog site. It embeds a copy of `registry.json` inside a `<script type="application/json" id="registry-data">` tag as a fallback for when the direct fetch fails.
-- `docs/script.js` fetches `registry.json` at runtime. If the fetch fails, it reads the embedded copy. It renders aggregate stats (total skills, rules, MCP tools) and tool cards with filtering by type and topic.
-- `docs/style.css` contains all styling -- dark theme, responsive layout, card design. No CSS framework.
+- All CSS and JS are inline in `docs/index.html`; there is no separate stylesheet or script file. The inline script fetches `registry.json` at runtime (falling back to the embedded copy) and adds search, category and status filters, and URL state on top of cards that `sync_from_registry.py` pre-renders.
+- Never edit the generated regions by hand (`registry:cards`, `registry:filters`, the embedded JSON blocks, hero stats); change `registry.json` and run `python scripts/sync_from_registry.py`.
 - **No external CDN dependencies.** The site must work fully offline except for badge images. No jQuery, no Bootstrap, no Google Fonts.
 - The `pages.yml` workflow copies `registry.json` and `assets/` into `docs/` at deploy time. You do not need to manually place these files in `docs/`.
 - When testing locally, open `docs/index.html` directly in a browser. The embedded fallback will kick in since `registry.json` won't be in `docs/` locally.
 
@@ -16,7 +16,7 @@ Developer Tools Directory is a **meta-repository** that catalogs, standardizes,
 ```
 Developer-Tools-Directory/
   registry.json              # Source of truth for all tool repos
-  standards/                 # 18 convention docs (structure, CI/CD, testing, MCP, security, licensing, etc.)
+  standards/                 # Convention docs (structure, CI/CD, testing, MCP, security, licensing, etc.)
   scaffold/
     create-tool.py           # Python repo generator (Jinja2)
     templates/               # Jinja2 templates for new repos
@@ -28,9 +28,9 @@ Developer-Tools-Directory/
     fonts/                   # Self-hosted Inter + JetBrains Mono woff2
     SETUP-PROMPT.md          # Copy-paste prompt for applying template to a repo
   docs/
-    index.html               # GitHub Pages catalog site
-    style.css                # Dark theme, responsive, card layout
-    script.js                # Fetches registry.json, renders cards, filters
+    index.html               # GitHub Pages catalog site (inline CSS and JS)
+    search-index.json        # Skill/rule/MCP-tool names for search (aggregate_search.py)
+    robots.txt, sitemap.xml  # Generated by sync_from_registry.py
     .well-known/security.txt # RFC 9116 security contact
     contributors/            # Worked examples for new contributors
   assets/
@@ -47,7 +47,7 @@ Developer-Tools-Directory/
 
 ### registry.json
 
-Array of tool objects. Each entry has: `name`, `repo`, `slug`, `description`, `type` (cursor-plugin | mcp-server), `homepage`, `skills`, `rules`, `mcpTools`, `extras`, `topics`, `status`, `language`, `license`, `pagesType`, `hasCI`. `status` is one of `experimental`, `beta`, `active`, `maintenance`, `deprecated`, or `archived` (see [`standards/lifecycle.md`](standards/lifecycle.md)).
+Array of tool objects. Each entry has: `name`, `repo`, `slug`, `description`, `type` (cursor-plugin | mcp-server), `category` (catalog filter chip; allowed values in `scripts/validate_registry.py`), `homepage`, `skills`, `rules`, `mcpTools`, `extras`, `topics`, `status`, `language`, `license`, `pagesType`, `hasCI`. `status` is one of `experimental`, `beta`, `active`, `maintenance`, `deprecated`, or `archived` (see [`standards/lifecycle.md`](standards/lifecycle.md)).
 
 There is intentionally no per-tool `version` field. Tool repos auto-bump patch on every content merge, so a catalog-pinned patch version re-stales on each tool's next content PR; the catalog tracks tools by repo and homepage instead. Do not reintroduce it.
 
@@ -67,11 +67,11 @@ Templates in `scaffold/templates/` produce: GitHub Actions workflows, README.md,
 
 ### docs/ (catalog site)
 
-Static HTML/CSS/JS. No build step. No external CDN. The `pages.yml` workflow copies `registry.json` and `assets/` into `docs/` at deploy time. `script.js` fetches `registry.json` at runtime and falls back to an embedded copy. Never use `innerHTML`/`eval` with registry data - the safety-scan CI job blocks reintroduction.
+Static single page; all CSS and JS are inline in `docs/index.html` (there is no separate stylesheet or script file). No build step. No external CDN. `sync_from_registry.py` pre-renders the tool cards, filter chips, and hero stats so the page works without JavaScript; the inline script fetches `registry.json` at runtime (falling back to the embedded copy) and layers search and filtering on top. The `pages.yml` workflow copies `registry.json` and `assets/` into `docs/` and stamps the footer date at deploy time. Refresh search data with `python site-template/aggregate_search.py --clone`. Never use `innerHTML`/`eval` with registry data - the safety-scan CI job blocks reintroduction.
 
 ### standards/
 
-17 Markdown documents defining conventions for every tool repo. Index in [`standards/README.md`](standards/README.md).
+Markdown documents defining conventions for every tool repo. The index in [`standards/README.md`](standards/README.md) is the authoritative list (and the catalog site renders a card per file).
 
 ## Cataloged Tools
 
@@ -180,66 +180,3 @@ Open `docs/index.html` in a browser. The script falls back to embedded registry
 | Jinja2 >=3.1,<4.0 | Scaffold template rendering | `requirements.txt` |
 
 The docs site has zero runtime dependencies. The sync script is pure stdlib.
-
-# context-mode — MANDATORY routing rules
-
-You have context-mode MCP tools available. These rules are NOT optional — they protect your context window from flooding. A single unrouted command can dump 56 KB into context and waste the entire session.
-
-## BLOCKED commands — do NOT attempt these
-
-### curl / wget — BLOCKED
-Any Bash command containing `curl` or `wget` is intercepted and replaced with an error message. Do NOT retry.
-Instead use:
-- `ctx_fetch_and_index(url, source)` to fetch and index web pages
-- `ctx_execute(language: "javascript", code: "const r = await fetch(...)")` to run HTTP calls in sandbox
-
-### Inline HTTP — BLOCKED
-Any Bash command containing `fetch('http`, `requests.get(`, `requests.post(`, `http.get(`, or `http.request(` is intercepted and replaced with an error message. Do NOT retry with Bash.
-Instead use:
-- `ctx_execute(language, code)` to run HTTP calls in sandbox — only stdout enters context
-
-### WebFetch — BLOCKED
-WebFetch calls are denied entirely. The URL is extracted and you are told to use `ctx_fetch_and_index` instead.
-Instead use:
-- `ctx_fetch_and_index(url, source)` then `ctx_search(queries)` to query the indexed content
-
-## REDIRECTED tools — use sandbox equivalents
-
-### Bash (>20 lines output)
-Bash is ONLY for: `git`, `mkdir`, `rm`, `mv`, `cd`, `ls`, `npm install`, `pip install`, and other short-output commands.
-For everything else, use:
-- `ctx_batch_execute(commands, queries)` — run multiple commands + search in ONE call
-- `ctx_execute(language: "shell", code: "...")` — run in sandbox, only stdout enters context
-
-### Read (for analysis)
-If you are reading a file to **Edit** it → Read is correct (Edit needs content in context).
-If you are reading to **analyze, explore, or summarize** → use `ctx_execute_file(path, language, code)` instead. Only your printed summary enters context. The raw file content stays in the sandbox.
-
-### Grep (large results)
-Grep results can flood context. Use `ctx_execute(language: "shell", code: "grep ...")` to run searches in sandbox. Only your printed summary enters context.
-
-## Tool selection hierarchy
-
-1. **GATHER**: `ctx_batch_execute(commands, queries)` — Primary tool. Runs all commands, auto-indexes output, returns search results. ONE call replaces 30+ individual calls.
-2. **FOLLOW-UP**: `ctx_search(queries: ["q1", "q2", ...])` — Query indexed content. Pass ALL questions as array in ONE call.
-3. **PROCESSING**: `ctx_execute(language, code)` | `ctx_execute_file(path, language, code)` — Sandbox execution. Only stdout enters context.
-4. **WEB**: `ctx_fetch_and_index(url, source)` then `ctx_search(queries)` — Fetch, chunk, index, query. Raw HTML never enters context.
-5. **INDEX**: `ctx_index(content, source)` — Store content in FTS5 knowledge base for later search.
-
-## Subagent routing
-
-When spawning subagents (Agent/Task tool), the routing block is automatically injected into their prompt. Bash-type subagents are upgraded to general-purpose so they have access to MCP tools. You do NOT need to manually instruct subagents about context-mode.
-
-## Output constraints
-
-- Keep responses under 500 words.
-- Write artifacts (code, configs, PRDs) to FILES — never return them as inline text. Return only: file path + 1-line description.
-- When indexing content, use descriptive source labels so others can `ctx_search(source: "label")` later.
-
-## ctx commands
-
-| Command | Action |
-|---------|--------|
-| `ctx stats` | Call the `ctx_stats` MCP tool and display the full output verbatim |
-| `ctx doctor` | Call the `ctx_doctor` MCP tool, run the returned shell command, display as checklist |
-| `ctx upgrade` | Call the `ctx_upgrade` MCP tool, run the returned shell command, display as checklist |
@@ -24,7 +24,7 @@ Thank you for your interest in contributing. This guide covers everything you ne
 ```
 Developer-Tools-Directory/
   registry.json              # Tool catalog (source of truth)
-  standards/                 # Convention documentation (18 docs)
+  standards/                 # Convention documentation (see standards/README.md)
   scaffold/
     create-tool.py           # Repo generator script
     templates/               # Jinja2 templates for new repos
@@ -48,15 +48,14 @@ Developer-Tools-Directory/
   "slug": "your-tool-repo",
   "description": "One-line description of the tool",
   "type": "cursor-plugin",
+  "category": "developer-tooling",
   "homepage": "https://tmhsdigital.github.io/Your-Tool-Repo/",
   "skills": 0,
   "rules": 0,
   "mcpTools": 0,
   "extras": {},
-  "npm": "",
   "topics": ["topic-1", "topic-2"],
   "status": "active",
-  "version": "0.1.0",
   "language": "Python",
   "license": "CC-BY-NC-ND-4.0",
   "pagesType": "static",
@@ -198,9 +197,8 @@ Templates use Jinja2. Available context variables:
 
 The catalog site lives in `docs/` and is vanilla HTML/CSS/JS with zero dependencies.
 
-- `docs/index.html` - page structure, embedded registry fallback
-- `docs/style.css` - dark theme, responsive cards
-- `docs/script.js` - fetches registry.json, renders stats and tool cards, filtering
+- `docs/index.html` - the whole site: markup, inline CSS, and inline JS (search, filters, URL state)
+- Generated regions (tool cards, filter chips, hero stats, embedded registry and search JSON) come from `scripts/sync_from_registry.py`; do not edit them by hand
 
 The `pages.yml` workflow copies `registry.json` and `assets/` into `docs/` at deploy time, so you don't need to manually copy files.
 
 
@@ -1,6 +1,6 @@
 # Performance
 
-> Lighthouse audit results for all TMHSDigital developer tool sites.
+> Historical baseline: a Lighthouse audit from 2026-04-09 covering the five tool sites that existed then. It predates the v1.4.0 site changes and the catalog's growth to 17 tools, so treat the numbers as a reference point, not current scores. Re-run Lighthouse against the live sites before relying on them.
 
 ## Audit Date
 
 
@@ -214,7 +214,7 @@ Developer-Tools-Directory/
     template.html.j2       Jinja2 HTML template with configurable branding
     fonts/                 Self-hosted Inter + JetBrains Mono woff2
     SETUP-PROMPT.md        Copy-paste prompt for applying template to a repo
-  standards/               Convention documentation (9 docs)
+  standards/               Convention documentation (index in standards/README.md)
   registry.json            Tool registry (source of truth)
   AGENTS.md                AI agent guidance
   CLAUDE.md                Claude Code project docs