@utilix-tech/mcp 0.81.0 → 0.83.0

package/CHANGELOG.md

@@ -2,6 +2,13 @@
 
 Full release notes for all Utilix surfaces (browser tools, REST API, SDKs, MCP server) live at [utilix.tech/changelog](https://utilix.tech/changelog). This file tracks just this package.
 
+## 0.83.0
+- Add `extract_markdown_link_references` tool: extract reference-style Markdown links (`[text][ref]`, the `[text][]` shortcut, and `![alt][ref]` image references) and match them against their `[ref]: url` definitions, flagging unresolved references and unused definitions.
+- Add `extract_prompt_template_variables` tool: extract `{{variable}}`-style placeholders from a prompt template, report occurrence counts, flag duplicates, and optionally cross-check against a list of provided variable names.
+
+## 0.82.0
+- Add `generate_robots_txt` tool: build a robots.txt file from User-agent groups with Allow/Disallow rules and an optional Crawl-delay, plus a list of Sitemap URLs. The generator-side sibling of `validate_robots_txt`.
+
 ## 0.81.0
 - Add `validate_fence_language_tags` tool: scan a Markdown document's fenced code blocks and flag a missing, unknown, or inconsistently-cased language tag against a curated list of common syntax-highlighter identifiers.
 - Add `count_sql_placeholders` tool: count positional (`?`, `$1`) and named (`:name`, `@name`) parameter placeholders in a SQL query, flagging mixed styles and numbered-sequence gaps or duplicates.

package/README.md

@@ -1,6 +1,6 @@
 # @utilix-tech/mcp
 
-MCP server exposing 257 developer utility tools to Claude, Cursor, VS Code Copilot, and any [Model Context Protocol](https://modelcontextprotocol.io) compatible AI assistant.
+MCP server exposing 260 developer utility tools to Claude, Cursor, VS Code Copilot, and any [Model Context Protocol](https://modelcontextprotocol.io) compatible AI assistant.
 
 No API key required. All tools run locally. Requires Node.js 18 or later.
 
@@ -231,6 +231,7 @@ claude mcp add utilix -- npx -y @utilix-tech/mcp
 | `validate_sitemap` | Validate a sitemap.xml or sitemap index against the sitemaps.org protocol |
 | `split_sitemap` | Split a URL list exceeding the sitemaps.org 50,000-URL cap into multiple sitemap.xml files plus a sitemap index file |
 | `validate_robots_txt` | Validate a robots.txt file: User-agent groups, rules, and directive syntax |
+| `generate_robots_txt` | Generate a robots.txt file from User-agent groups with Allow/Disallow rules and an optional Crawl-delay, plus Sitemap URLs |
 | `parse_user_agent` | Parse a browser User-Agent string into browser, engine, OS, and device type |
 | `build_cache_control` | Build and explain an HTTP Cache-Control response header from directives (max-age, no-store, immutable, etc.) |
 | `build_referrer_policy` | Build and explain an HTTP Referrer-Policy header from one or more fallback-chain policy tokens |
@@ -281,7 +282,7 @@ claude mcp add utilix -- npx -y @utilix-tech/mcp
 | `color_palette` | Generate a color palette from a base color |
 | `color_shades` | Generate shades or tints from a base color |
 
-### Code (24 tools)
+### Code (25 tools)
 | Tool | Description |
 |------|-------------|
 | `format_sql` | Format a SQL query with proper indentation |
@@ -308,6 +309,7 @@ claude mcp add utilix -- npx -y @utilix-tech/mcp
 | `parse_connection_string` | Parse a database connection URI (postgres, mysql, mongodb, redis, amqp, and similar) into its scheme, credentials, hosts, database, and query parameters |
 | `validate_fence_language_tags` | Scan a Markdown document's fenced code blocks and flag a missing, unknown, or inconsistently-cased language tag against a curated list of common syntax-highlighter identifiers |
 | `count_sql_placeholders` | Count positional (`?`, `$1`) and named (`:name`, `@name`) parameter placeholders in a SQL query, flagging mixed styles and numbered-sequence gaps or duplicates |
+| `extract_markdown_link_references` | Extract every reference-style Markdown link ([text][ref], [text][], ![alt][ref]) and match it against its [ref]: url definition, flagging unresolved references and unused definitions |
 
 ### Data (16 tools)
 | Tool | Description |
@@ -343,7 +345,7 @@ claude mcp add utilix -- npx -y @utilix-tech/mcp
 | `unicode_inspect` | Inspect Unicode code points of a string |
 | `optimize_svg` | Optimize and minify an SVG string |
 
-### AI / Agent (29 tools)
+### AI / Agent (30 tools)
 | Tool | Description |
 |------|-------------|
 | `estimate_tokens` | Estimate token count and cost for an LLM prompt |
@@ -375,6 +377,7 @@ claude mcp add utilix -- npx -y @utilix-tech/mcp
 | `format_few_shot_examples` | Format input/output example pairs into a numbered, XML-tagged, or delimiter-separated few-shot prompt template |
 | `lint_tool_schema` | Lint an OpenAI, Anthropic, or MCP tool/function-calling definition for missing descriptions, invalid names, empty enums, and required/properties mismatches |
 | `calculate_retry_backoff_schedule` | Compute the full sequence of delays an exponential-backoff retry policy would actually produce, given a base delay, multiplier, and max retry count, optionally capped at a max delay and with full/equal jitter ranges |
+| `extract_prompt_template_variables` | Extract every {{variable}}-style placeholder from a prompt template, report occurrence counts, flag duplicates, and optionally cross-check against a list of provided variable names |
 
 ### Image (4 tools)
 | Tool | Description |

package/dist/index.js

@@ -1895,6 +1895,58 @@ server.tool(
     return text(JSON.stringify(result, null, 2));
   }
 );
+function _generateRobotsTxt(groups, sitemaps) {
+  if (!Array.isArray(groups) || groups.length === 0) {
+    return { error: "At least one group with a user-agent is required" };
+  }
+  const lines = [];
+  for (const group of groups) {
+    const userAgents = (group.userAgents ?? []).map((a) => a.trim()).filter(Boolean);
+    if (userAgents.length === 0) return { error: "Each group needs at least one non-empty user-agent" };
+    for (const ua of userAgents) lines.push(`User-agent: ${ua}`);
+    const disallow = group.disallow ?? [];
+    const allow = group.allow ?? [];
+    if (disallow.length === 0 && allow.length === 0) {
+      lines.push("Disallow:");
+    } else {
+      for (const value of disallow) lines.push(`Disallow: ${value}`);
+      for (const value of allow) lines.push(`Allow: ${value}`);
+    }
+    if (group.crawlDelay !== void 0) {
+      if (!Number.isFinite(group.crawlDelay) || group.crawlDelay < 0) {
+        return { error: "crawlDelay must be a non-negative number" };
+      }
+      lines.push(`Crawl-delay: ${group.crawlDelay}`);
+    }
+    lines.push("");
+  }
+  for (const sitemap of sitemaps ?? []) {
+    if (!/^https?:\/\//i.test(sitemap)) {
+      return { error: `Sitemap URL "${sitemap}" must be absolute (start with http:// or https://)` };
+    }
+    lines.push(`Sitemap: ${sitemap}`);
+  }
+  while (lines.length > 0 && lines[lines.length - 1] === "") lines.pop();
+  return { text: lines.join("\n") + "\n" };
+}
+server.tool(
+  "generate_robots_txt",
+  "Generate a robots.txt file from one or more User-agent groups (each with Allow/Disallow rules and an optional Crawl-delay) plus a list of Sitemap URLs. The inverse of validate_robots_txt: a group with no Allow or Disallow rules gets an explicit empty Disallow line, the standard way to allow crawling everything.",
+  {
+    groups: z.array(z.object({
+      userAgents: z.array(z.string()).describe('User-agent tokens for this group, e.g. ["*"] or ["Googlebot"]'),
+      allow: z.array(z.string()).optional().describe('Allow rule paths, e.g. ["/admin/public"]'),
+      disallow: z.array(z.string()).optional().describe('Disallow rule paths, e.g. ["/admin"]'),
+      crawlDelay: z.number().optional().describe("Crawl-delay in seconds for this group")
+    })).describe("User-agent groups to emit, in order"),
+    sitemaps: z.array(z.string()).optional().describe("Absolute sitemap URLs to append as Sitemap directives")
+  },
+  async ({ groups, sitemaps }) => {
+    const result = _generateRobotsTxt(groups, sitemaps);
+    if ("error" in result) return errText(result);
+    return text(JSON.stringify(result, null, 2));
+  }
+);
 var _CARD_NETWORK_RULES = [
   { name: "American Express", lengths: [15], test: (d) => /^3[47]/.test(d) },
   { name: "Diners Club", lengths: [14], test: (d) => /^3(?:0[0-5]|[68])/.test(d) },
@@ -13107,5 +13159,117 @@ server.tool(
     return text(JSON.stringify(result, null, 2));
   }
 );
+var _LINK_REF_DEFINITION_REGEX_MCP = /^ {0,3}\[([^\]]+)\]:\s*(\S+)(?:\s+(?:"([^"]*)"|'([^']*)'|\(([^)]*)\)))?\s*$/;
+var _LINK_REF_USAGE_REGEX_MCP = /(!?)\[([^[\]\n]*)\]\[([^[\]\n]*)\]/g;
+function _normalizeLinkLabelMcp(label) {
+  return label.trim().replace(/\s+/g, " ").toLowerCase();
+}
+function _extractMarkdownLinkReferencesMcp(markdown) {
+  const trimmed = markdown.trim();
+  if (!trimmed) return { error: "Empty input" };
+  const lines = markdown.replace(/\r\n/g, "\n").split("\n");
+  const definitions = [];
+  const definitionsByLabel = /* @__PURE__ */ new Map();
+  for (const line of lines) {
+    const match = line.match(_LINK_REF_DEFINITION_REGEX_MCP);
+    if (!match) continue;
+    const label = match[1];
+    const normalized = _normalizeLinkLabelMcp(label);
+    if (definitionsByLabel.has(normalized)) continue;
+    const def = { label, url: match[2], title: match[3] ?? match[4] ?? match[5] ?? void 0 };
+    definitionsByLabel.set(normalized, def);
+    definitions.push(def);
+  }
+  const usages = [];
+  for (const m of markdown.matchAll(_LINK_REF_USAGE_REGEX_MCP)) {
+    const image = m[1] === "!";
+    const text2 = m[2];
+    const rawRef = m[3];
+    const label = rawRef.trim() === "" ? text2 : rawRef;
+    usages.push({ text: text2, label, image });
+  }
+  if (definitions.length === 0 && usages.length === 0) {
+    return { error: "No reference-style links or definitions found" };
+  }
+  const usedLabels = /* @__PURE__ */ new Set();
+  const resolved = [];
+  const unresolvedReferences = [];
+  for (const usage of usages) {
+    const normalized = _normalizeLinkLabelMcp(usage.label);
+    const def = definitionsByLabel.get(normalized);
+    if (def) {
+      usedLabels.add(normalized);
+      resolved.push({ ...usage, url: def.url, title: def.title });
+    } else {
+      unresolvedReferences.push(usage);
+    }
+  }
+  const unusedDefinitions = definitions.filter((d) => !usedLabels.has(_normalizeLinkLabelMcp(d.label)));
+  return {
+    referenceCount: usages.length,
+    definitionCount: definitions.length,
+    resolved,
+    unresolvedReferences,
+    unusedDefinitions
+  };
+}
+server.tool(
+  "extract_markdown_link_references",
+  'Extract every reference-style Markdown link ([text][ref], including the shortcut form [text][] and image references ![alt][ref]) alongside its [ref]: url "title" definition, then flag usages with no matching definition and definitions that no usage ever references.',
+  {
+    markdown: z.string().min(1).describe("The Markdown document text")
+  },
+  async ({ markdown }) => {
+    const result = _extractMarkdownLinkReferencesMcp(markdown);
+    if ("error" in result) return errText(result);
+    return text(JSON.stringify(result, null, 2));
+  }
+);
+var _PROMPT_TEMPLATE_VARIABLE_REGEX_MCP = /\{\{\s*([a-zA-Z0-9_.-]+)\s*\}\}/g;
+function _extractPromptTemplateVariablesMcp(template, providedVars) {
+  const trimmed = template.trim();
+  if (!trimmed) return { error: "Empty input" };
+  const order = [];
+  const counts = /* @__PURE__ */ new Map();
+  for (const m of template.matchAll(_PROMPT_TEMPLATE_VARIABLE_REGEX_MCP)) {
+    const name = m[1];
+    if (!counts.has(name)) {
+      counts.set(name, 0);
+      order.push(name);
+    }
+    counts.set(name, (counts.get(name) ?? 0) + 1);
+  }
+  if (order.length === 0) return { error: "No {{variable}} placeholders found" };
+  const variables = order.map((name) => ({ name, count: counts.get(name) ?? 0 }));
+  const duplicates = order.filter((name) => (counts.get(name) ?? 0) > 1);
+  let missing = [];
+  const unused = [];
+  if (providedVars !== void 0) {
+    const providedSet = new Set(providedVars.map((v) => v.trim()).filter(Boolean));
+    const usedSet = new Set(order);
+    missing = order.filter((name) => !providedSet.has(name));
+    const seenProvided = /* @__PURE__ */ new Set();
+    for (const raw of providedVars) {
+      const v = raw.trim();
+      if (!v || usedSet.has(v) || seenProvided.has(v)) continue;
+      seenProvided.add(v);
+      unused.push(v);
+    }
+  }
+  return { variableCount: order.length, variables, duplicates, missing, unused };
+}
+server.tool(
+  "extract_prompt_template_variables",
+  "Extract every {{variable}}-style placeholder from a prompt template, reporting each unique variable's occurrence count and flagging ones used more than once. Optionally cross-check against a providedVars list to flag variables the template uses that are missing, and provided names the template never references.",
+  {
+    template: z.string().min(1).describe("The prompt template text containing {{variable}} placeholders"),
+    providedVars: z.array(z.string()).optional().describe("Optional list of variable names already available, to cross-check against the template")
+  },
+  async ({ template, providedVars }) => {
+    const result = _extractPromptTemplateVariablesMcp(template, providedVars);
+    if ("error" in result) return errText(result);
+    return text(JSON.stringify(result, null, 2));
+  }
+);
 var transport = new StdioServerTransport();
 await server.connect(transport);

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@utilix-tech/mcp",
-  "version": "0.81.0",
+  "version": "0.83.0",
   "description": "MCP server exposing 198 Utilix developer tools to Claude, Cursor, VS Code Copilot, and any MCP-compatible AI assistant.",
   "author": "Utilix <hello@utilix.tech>",
   "license": "MIT",