@gleanwork/mcp-server-tester 1.0.0-beta.8 → 1.0.1-beta.0

package/README.md

@@ -126,7 +126,7 @@ LLM host mode makes real API calls and produces non-deterministic results. Use `
 Requires Node.js 22+.
 
 ```bash
-npm install --save-dev @gleanwork/mcp-server-tester @playwright/test zod
+npm install --save-dev @gleanwork/mcp-server-tester @playwright/test
 ```
 
 The Anthropic SDK is only needed for LLM-as-judge assertions or LLM host mode with the Anthropic provider:
@@ -182,6 +182,25 @@ For HTTP servers, set `transport: 'http'` and `serverUrl`. For servers that requ
 - [Development](./docs/development.md) — contributing and building
 - [Migration Guide (v0.12 → v1.0)](./docs/migrations/migration-1.0.md) — upgrading from pre-1.0 releases
 
+## AI Skills
+
+Install AI skills to help your coding assistant generate tests, eval datasets, and MCP host evals:
+
+```bash
+npx skills add -g gleanwork/mcp-server-tester
+```
+
+This installs skills globally so they're available across all your projects. Four skills are included:
+
+| Skill                 | Description                                                 |
+| --------------------- | ----------------------------------------------------------- |
+| `mcp-tester-guide`    | Framework reference — matchers, config, auth, anti-patterns |
+| `write-mcp-test`      | Generate direct-mode Playwright tests                       |
+| `write-mcp-eval`      | Generate data-driven eval datasets                          |
+| `write-mcp-host-eval` | Generate LLM host simulation evals                          |
+
+Compatible with Claude Code, Cursor, Windsurf, Copilot, and [40+ other AI agents](https://github.com/nicepkg/nice-skills).
+
 ## Examples
 
 The `examples/` directory contains complete working examples:

package/dist/cli/index.js

@@ -80,7 +80,7 @@ function JsonPreview({ data, maxLines = 15 }) {
 
 // package.json
 var package_default = {
-  version: "1.0.0-beta.8"};
+  version: "1.0.1-beta.0"};
 
 // src/cli/templates/index.ts
 function getPlaywrightConfigTemplate(answers) {
@@ -1005,6 +1005,17 @@ async function createMCPClientForConfig(config, options) {
 }
 async function closeMCPClient(client) {
   try {
+    const transport = client.transport;
+    if (transport instanceof StreamableHTTPClientTransport) {
+      try {
+        await transport.terminateSession();
+      } catch (sessionError) {
+        debugClient(
+          "Error terminating session: %s",
+          sessionError instanceof Error ? sessionError.message : String(sessionError)
+        );
+      }
+    }
     await client.close();
   } catch (error) {
     debugClient(

package/dist/fixtures/mcp.js

@@ -237,11 +237,13 @@ function validateSchema(response, schema, options = {}) {
   } catch (error) {
     const zodError = error;
     const issues = formatZodIssues(zodError);
+    const text = stringifyResponse(response);
     return {
       pass: false,
       message: `Response does not match schema: ${issues}`,
       details: {
-        issues: zodError.issues
+        issues: zodError.issues,
+        textPreview: truncateForDisplay2(text)
       }
     };
   }
@@ -294,16 +296,29 @@ function formatZodIssues(error) {
   });
   return issues.join("; ");
 }
+function truncateForDisplay2(str, maxLength = 200) {
+  if (str.length <= maxLength) {
+    return str;
+  }
+  return str.slice(0, maxLength) + "... (truncated)";
+}
 
 // src/assertions/matchers/toMatchToolSchema.ts
 function toMatchToolSchema(received, schema, options = {}) {
   const result = validateSchema(received, schema, options);
+  const preview = result.details?.textPreview;
   return {
     pass: result.pass,
     message: () => {
       if (this.isNot) {
         return result.pass ? "Expected response NOT to match schema, but it did" : result.message;
       }
+      if (!result.pass && preview) {
+        return `${result.message}
+
+Actual response (truncated):
+${preview}`;
+      }
       return result.message;
     }
   };
@@ -334,11 +349,11 @@ function validateText(response, expected, options = {}) {
     details: {
       missing,
       textLength: text.length,
-      textPreview: truncateForDisplay2(text)
+      textPreview: truncateForDisplay3(text)
     }
   };
 }
-function truncateForDisplay2(str, maxLength = 200) {
+function truncateForDisplay3(str, maxLength = 200) {
   if (str.length <= maxLength) {
     return str;
   }
@@ -348,6 +363,7 @@ function truncateForDisplay2(str, maxLength = 200) {
 // src/assertions/matchers/toContainToolText.ts
 function toContainToolText(received, expected, options = {}) {
   const result = validateText(received, expected, options);
+  const preview = result.details?.textPreview;
   return {
     pass: result.pass,
     message: () => {
@@ -355,6 +371,12 @@ function toContainToolText(received, expected, options = {}) {
         const expectedStr = Array.isArray(expected) ? expected.map((s) => `"${s}"`).join(", ") : `"${expected}"`;
         return result.pass ? `Expected response NOT to contain ${expectedStr}, but it did` : result.message;
       }
+      if (!result.pass && preview) {
+        return `${result.message}
+
+Actual response (truncated):
+${preview}`;
+      }
       return result.message;
     }
   };
@@ -385,7 +407,7 @@ function validatePattern(response, patterns, options = {}) {
     details: {
       unmatched,
       textLength: text.length,
-      textPreview: truncateForDisplay3(text)
+      textPreview: truncateForDisplay4(text)
     }
   };
 }
@@ -405,7 +427,7 @@ function patternToString(pattern) {
   }
   return `/${pattern}/`;
 }
-function truncateForDisplay3(str, maxLength = 200) {
+function truncateForDisplay4(str, maxLength = 200) {
   if (str.length <= maxLength) {
     return str;
   }
@@ -415,12 +437,19 @@ function truncateForDisplay3(str, maxLength = 200) {
 // src/assertions/matchers/toMatchToolPattern.ts
 function toMatchToolPattern(received, patterns, options = {}) {
   const result = validatePattern(received, patterns, options);
+  const preview = result.details?.textPreview;
   return {
     pass: result.pass,
     message: () => {
       if (this.isNot) {
         return result.pass ? "Expected response NOT to match pattern(s), but it did" : result.message;
       }
+      if (!result.pass && preview) {
+        return `${result.message}
+
+Actual response (truncated):
+${preview}`;
+      }
       return result.message;
     }
   };
@@ -567,7 +596,7 @@ function validateError(response, expected = true) {
         pass: false,
         message: "Expected an error response but got success",
         details: {
-          textPreview: truncateForDisplay4(extractText2(response))
+          textPreview: truncateForDisplay5(extractText2(response))
         }
       };
     } else {
@@ -579,7 +608,7 @@ function validateError(response, expected = true) {
       }
       return {
         pass: false,
-        message: `Expected a success response but got error: "${truncateForDisplay4(errorMessage)}"`,
+        message: `Expected a success response but got error: "${truncateForDisplay5(errorMessage)}"`,
         details: {
           errorMessage
         }
@@ -592,7 +621,7 @@ function validateError(response, expected = true) {
       pass: false,
       message: `Expected an error containing "${expectedMessages[0]}" but got success`,
       details: {
-        textPreview: truncateForDisplay4(extractText2(response))
+        textPreview: truncateForDisplay5(extractText2(response))
       }
     };
   }
@@ -614,7 +643,7 @@ function validateError(response, expected = true) {
     }
   };
 }
-function truncateForDisplay4(str, maxLength = 200) {
+function truncateForDisplay5(str, maxLength = 200) {
   if (str.length <= maxLength) {
     return str;
   }
@@ -1142,7 +1171,9 @@ function createJudge(config = {}) {
     case "google":
       return createGoogleJudge(config);
     default:
-      throw new Error(`Unsupported LLM provider: ${String(provider)}`);
+      throw new Error(
+        `Unsupported LLM provider: ${String(provider)}. Valid providers: 'anthropic', 'vertex-anthropic', 'anthropic-agent-sdk', 'openai', 'google'`
+      );
   }
 }
 
@@ -1497,6 +1528,10 @@ function validateToolCalls(response, expectation) {
           return {
             pass: false,
             message: `Expected tool '${expected.name}' to be called in sequence (starting from position ${searchFrom}), but it was not found`,
+            details: {
+              actual: actual.map((c) => c.name),
+              expected: expected.name
+            },
             metrics
           };
         }
@@ -1513,6 +1548,10 @@ function validateToolCalls(response, expectation) {
         return {
           pass: false,
           message: `Expected tool '${expected.name}'${argsNote} to be called, but it was not`,
+          details: {
+            actual: actual.map((c) => c.name),
+            expected: expected.name
+          },
           metrics
         };
       }
@@ -1525,6 +1564,10 @@ function validateToolCalls(response, expectation) {
       return {
         pass: false,
         message: `Unexpected tool calls: ${names}. Only ${[...allowedNames].map((n) => `'${n}'`).join(", ")} are allowed`,
+        details: {
+          actual: actual.map((c) => c.name),
+          unexpected: unexpected.map((c) => c.name)
+        },
         metrics
       };
     }
@@ -1543,19 +1586,22 @@ function validateToolCallCount(response, options) {
   if (exact !== void 0 && count !== exact) {
     return {
       pass: false,
-      message: `Expected exactly ${exact} tool call(s), but got ${count}`
+      message: `Expected exactly ${exact} tool call(s), but got ${count}`,
+      details: { actual: count, expected: exact }
     };
   }
   if (min !== void 0 && count < min) {
     return {
       pass: false,
-      message: `Expected at least ${min} tool call(s), but got ${count}`
+      message: `Expected at least ${min} tool call(s), but got ${count}`,
+      details: { actual: count, min }
     };
   }
   if (max !== void 0 && count > max) {
     return {
       pass: false,
-      message: `Expected at most ${max} tool call(s), but got ${count}`
+      message: `Expected at most ${max} tool call(s), but got ${count}`,
+      details: { actual: count, max }
     };
   }
   return {
@@ -1694,7 +1740,7 @@ var debugHttp = createDebug(`${NAMESPACE}:http`);
 
 // package.json
 var package_default = {
-  version: "1.0.0-beta.8"};
+  version: "1.0.1-beta.0"};
 var debug = createDebug("mcp-server-tester:oauth-flow");
 async function generatePKCE() {
   const codeVerifier = oauth.generateRandomCodeVerifier();
@@ -2075,6 +2121,17 @@ async function createMCPClientForConfig(config, options) {
 }
 async function closeMCPClient(client) {
   try {
+    const transport = client.transport;
+    if (transport instanceof StreamableHTTPClientTransport) {
+      try {
+        await transport.terminateSession();
+      } catch (sessionError) {
+        debugClient(
+          "Error terminating session: %s",
+          sessionError instanceof Error ? sessionError.message : String(sessionError)
+        );
+      }
+    }
     await client.close();
   } catch (error) {
     debugClient(