@skyramp/mcp 0.3.5 → 0.3.6-rc.2.ac20

package/build/adapters/jestAdapter.js

@@ -1,3 +1,4 @@
+import * as path from "path";
 /**
  * Shared jest + vitest adapter for skyramp_run_existing_tests. vitest's json
  * reporter is jest-compatible, so both use ONE parser; only the reporter flag
@@ -40,6 +41,7 @@ export function parseJestJson(report, opts) {
             results.push({
                 testId: `${name} › (file failed to run)`,
                 file: name,
+                ...(path.isAbsolute(name) ? { absoluteFile: name } : {}),
                 status: "error",
                 message: stripVTControlCharacters(raw).trim() || undefined,
                 durationMs: 0,
@@ -55,6 +57,7 @@ export function parseJestJson(report, opts) {
             results.push({
                 testId: `${name} › ${titlePath}`,
                 file: name,
+                ...(path.isAbsolute(name) ? { absoluteFile: name } : {}),
                 status,
                 message,
                 durationMs: Math.round(a.duration ?? 0),

package/build/adapters/mochaAdapter.js

@@ -1,3 +1,4 @@
+import * as path from "path";
 /**
  * Mocha adapter for skyramp_run_existing_tests. Normalizes `mocha --reporter
  * json` (emitted to stdout) into the neutral shape. Mocha does not tag
@@ -29,6 +30,7 @@ function toResult(t, status) {
     return {
         testId: `${file} › ${full}`,
         file,
+        ...(path.isAbsolute(file) ? { absoluteFile: file } : {}),
         status,
         message: status === "fail" || status === "error" ? messageOf(t) : undefined,
         durationMs: Math.round(t.duration ?? 0),

package/build/adapters/playwrightAdapter.js

@@ -12,6 +12,7 @@
  * builds the flags appended to the suite's `testRunCommand`. The IO shell that
  * actually spawns the run lives in `runExistingTestsTool`.
  */
+import * as path from "path";
 import { stripVTControlCharacters } from "util";
 /**
  * A Playwright project (or spec file) is "infra" when it exists to bring up /
@@ -67,6 +68,7 @@ function messageOf(test) {
 }
 export function parsePlaywrightJson(report, opts) {
     const rep = (report ?? {});
+    const rootDir = rep.config?.rootDir;
     const collected = [];
     for (const fileSuite of rep.suites ?? []) {
         collectSpecs(fileSuite, fileSuite.file ?? "", [], collected);
@@ -118,6 +120,7 @@ export function parsePlaywrightJson(report, opts) {
             results.push({
                 testId,
                 file,
+                ...(rootDir ? { absoluteFile: path.resolve(rootDir, file) } : {}),
                 status,
                 message: status === "fail" || status === "error" ? messageOf(test) : undefined,
                 durationMs,

package/build/adapters/pytestAdapter.js

@@ -1,3 +1,4 @@
+import * as path from "path";
 /**
  * pytest adapter for skyramp_run_existing_tests. Normalizes the
  * `pytest-json-report` (`--json-report`) output into the neutral result shape.
@@ -39,6 +40,8 @@ function durationMsOf(test) {
 export function parsePytestJson(report, opts) {
     const rep = (report ?? {});
     const tests = rep.tests ?? [];
+    const root = rep.root;
+    const abs = (f) => (root ? { absoluteFile: path.resolve(root, f) } : {});
     // Real test outcomes.
     const testResults = tests.map((t) => {
         const status = mapOutcome(t.outcome);
@@ -46,6 +49,7 @@ export function parsePytestJson(report, opts) {
         return {
             testId: nodeid,
             file: fileOf(nodeid),
+            ...abs(fileOf(nodeid)),
             status,
             message: status === "fail" || status === "error" ? messageOf(t) : undefined,
             durationMs: durationMsOf(t),
@@ -61,6 +65,7 @@ export function parsePytestJson(report, opts) {
         return {
             testId: `${nodeid} › (collection error)`,
             file: fileOf(nodeid),
+            ...abs(fileOf(nodeid)),
             status: "error",
             message: lastLine(c.longrepr),
             durationMs: 0,
@@ -72,6 +77,13 @@ export function parsePytestJson(report, opts) {
     // When collection failed AND no real test ran, the suite as a whole could not be
     // collected — environmental (a wall of red), not PR signal. When some real tests
     // DID run, the collection errors above stand as ordinary PR-signal error results.
+    // NOTE: a collection error names the file that failed to import, but says nothing
+    // about WHY. A missing settings module, a database that is down or any shared
+    // dependency failure surfaces per-module, landing on exactly the files the run
+    // selected — indistinguishable here from a file the PR itself broke. So a run that
+    // collected nothing stays unhealthy even when scoped, and the file keeps its
+    // Unknown baseline. Treating it as PR signal would let an environment outage be
+    // reported as a test the change broke, and then repaired.
     if (collectionErrors.length > 0 && testResults.length === 0) {
         environmentHealthy = false;
         const first = collectionErrors[0];

package/build/playwright/registerPlaywrightTools.js

@@ -4,36 +4,111 @@
  */
 import { z } from "zod";
 import { logger } from "../utils/logger.js";
-import { dualChannelResult } from "../utils/utils.js";
 import { PlaywrightTraceService, } from "./PlaywrightTraceService.js";
 let _service = null;
 export function getPlaywrightTraceService() {
     return _service;
 }
 /**
- * The one browser tool whose result is re-shaped on the way out — see
- * toStructuredBlueprintResult.
+ * The one browser tool whose text payload is shrunk on the way out — see
+ * shrinkBlueprintResult.
  */
 const BLUEPRINT_TOOL = "browser_blueprint";
 /**
- * Re-shape a browser_blueprint result onto both MCP result channels.
+ * Shrink a browser_blueprint result on the text channel it already arrives on.
  *
- * SKYR-4194. The browser tools are registered here as passthroughs, and the fork
- * returns the captured DOM as one `content[]` text block with no
- * `structuredContent` — the shape whose spill the agent's file reader refuses.
- * See `dualChannelResult` for why both channels carry the payload.
+ * SKYR-4267. SKYR-4194 (#752) also copied the capture into `structuredContent`
+ * as a JSON string. Claude Code measures that copy against its size cap and
+ * drops `content[]` when both channels are present, so the escape added to
+ * every quote and newline counted against the cap. On the largest real capture
+ * (`sb20-bucket-visibility`) the text is 26,542 bytes and the escaped string
+ * 29,411 — which BOTH channels carried, because `dualChannelResult` puts the
+ * JSON serialization of `structuredContent` into `content[]`, so 58,822 went on
+ * the wire.
  *
- * An error result is returned untouched: the SDK's outputSchema check returns
- * early on `isError`, so it needs no structuredContent.
+ * The escaping also collapsed the whole payload onto ONE line, so a spill file
+ * read at a line offset returned nothing at all
+ * (`repo14-U06-field-reorder-toast`). Dropping it does not give the capture its
+ * newlines back: the fork pretty-prints the JSON, but shrinkBlueprintText below
+ * re-serializes with no indent, so the body is still one line. What returns are
+ * the `### ` section headers around it, so an offset read now lands on a header
+ * instead of erroring — the failure is mitigated, not closed.
+ *
+ * The text channel is the one every agent CLI reads, and it is what the other
+ * browser tools already return, so the capture rides it alone.
+ *
+ * An error result is returned untouched.
  */
-function toStructuredBlueprintResult(result) {
+function shrinkBlueprintResult(result) {
     if (result?.isError)
         return result;
-    const blueprint = (result?.content ?? [])
-        .filter((block) => block?.type === "text")
-        .map((block) => block.text)
-        .join("\n");
-    return dualChannelResult({ blueprint });
+    return {
+        ...result,
+        content: (result?.content ?? []).map((block) => block?.type === "text"
+            ? { ...block, text: shrinkBlueprintText(block.text) }
+            : block),
+    };
+}
+/**
+ * Drop every null-valued key, at any depth.
+ *
+ * SKYR-4208. Array entries are left alone — a null there holds a position, and
+ * removing it would renumber everything after it.
+ */
+function dropNulls(value) {
+    if (Array.isArray(value))
+        return value.map(dropNulls);
+    if (value === null || typeof value !== "object")
+        return value;
+    return Object.fromEntries(Object.entries(value)
+        .filter(([, v]) => v !== null)
+        .map(([k, v]) => [k, dropNulls(v)]));
+}
+/**
+ * Minify the JSON body of a browser_blueprint payload and drop its null keys.
+ *
+ * SKYR-4208. The fork pretty-prints the capture behind a `### Result` header.
+ * On eval run 32458076607 (962 payloads, 11,368 elements) the indentation is
+ * 35.4% of the payload's tokens, and the null keys a further 6.8% — `stableId`,
+ * `testId` and `fingerprint` are null on 72-94% of elements and cost their key
+ * name on every one.
+ *
+ * Both steps are lossless: the agent parses the same fields with the same
+ * values, and an absent key already says what a null key says. Keys that merely
+ * hold their DEFAULT (`framePath: []`, `shadowRoot: false`) are deliberately
+ * kept — dropping those makes the agent infer the default from an absence, and
+ * `framePath` stops being empty as soon as the page has an iframe.
+ *
+ * The header and anything else outside the JSON survive unchanged, and a body
+ * that does not parse is returned as it arrived.
+ */
+function shrinkBlueprintText(text) {
+    const start = text.indexOf("{");
+    if (start === -1)
+        return text;
+    // The fork can append more sections after `### Result` — `### Page` whenever
+    // the title or URL changed since the last call, and `### Modal state`,
+    // `### Snapshot` or `### Events` when the page produced them. Only the Result
+    // section holds JSON, so the body has to stop at the next header; reading to
+    // the end of the string makes the parse throw and costs the whole saving.
+    // A `### ` at the start of a line cannot sit inside the JSON, because JSON
+    // escapes the newline that would have to precede it.
+    const nextSection = text.slice(start).search(/\n### /);
+    const end = nextSection === -1 ? text.length : start + nextSection;
+    // The blank line the fork puts after a section belongs to the payload, not to
+    // the JSON, so it is put back byte-for-byte rather than parsed away.
+    const body = text.slice(start, end);
+    const json = body.trimEnd();
+    try {
+        return (text.slice(0, start) +
+            JSON.stringify(dropNulls(JSON.parse(json))) +
+            body.slice(json.length) +
+            text.slice(end));
+    }
+    catch {
+        logger.warning("browser_blueprint payload did not parse as JSON — delivering it unshrunk (SKYR-4208)");
+        return text;
+    }
 }
 export async function registerPlaywrightTools(server, options) {
     _service = new PlaywrightTraceService(options);
@@ -98,19 +173,6 @@ export async function registerPlaywrightTools(server, options) {
         server.registerTool(tool.name, {
             description: tool.description || `Playwright tool: ${tool.name}`,
             inputSchema: zodSchema,
-            // SKYR-4194: only browser_blueprint returns its payload via
-            // structuredContent, which requires a declared outputSchema. The other
-            // browser tools return small results, and some return images, which this
-            // text-only re-shaping would drop.
-            ...(isBlueprint
-                ? {
-                    outputSchema: {
-                        blueprint: z
-                            .string()
-                            .describe("The captured PageBlueprint — sections, elements, and their logical names. Read and follow this field."),
-                    },
-                }
-                : {}),
         }, async (params) => {
             if (!_service?.isInitialized()) {
                 return {
@@ -124,7 +186,7 @@ export async function registerPlaywrightTools(server, options) {
                 };
             }
             const result = await _service.callTool(tool.name, params);
-            return isBlueprint ? toStructuredBlueprintResult(result) : result;
+            return isBlueprint ? shrinkBlueprintResult(result) : result;
         });
     }
     logger.info(`Registered ${filteredTools.length} Playwright tools: ${filteredTools.map((t) => t.name).join(", ")}`);

package/build/playwright/traceRecordingPrompt.d.ts

@@ -8,6 +8,12 @@ export interface TraceRecordingPromptOpts {
     outputDir?: string;
     /** Whether to run skyramp_modularization after generation. Default: true. Set false in CI. */
     modularize?: boolean;
+    /**
+     * With modularize=false: the generation result itself may still instruct a
+     * modularization pass (codeReuse flows that seed a shared utils file). When
+     * true, the note defers to that result instead of banning the tool outright.
+     */
+    modularizeViaGenerationResult?: boolean;
 }
 export declare function getTraceRecordingPromptText(opts?: TraceRecordingPromptOpts): string;
 export declare function registerTraceRecordingPrompt(server: McpServer): void;

package/build/playwright/traceRecordingPrompt.js

@@ -13,10 +13,14 @@ export function getTraceRecordingPromptText(opts) {
         : `Call \`skyramp_export_zip\` with \`outputPath\` set to the absolute zip path (same directory and base name as the test file, replacing \`.spec.ts\` with \`.zip\`).`;
     const generateInstruction = modularize
         ? `Call \`skyramp_ui_test_generation\` with \`playwrightInput\` set to the absolute zip path from the Export step.`
-        : `Call \`skyramp_ui_test_generation\` with \`playwrightInput\` set to the absolute zip path from step 5 and \`modularizeCode: false\`.`;
+        : opts?.modularizeViaGenerationResult
+            ? `Call \`skyramp_ui_test_generation\` with \`playwrightInput\` set to the absolute zip path from step 5, \`modularizeCode: false\`, and \`codeReuse: true\` — its result lists the post-generation steps (assertion enhancement, modularization, reuse) with their arguments.`
+            : `Call \`skyramp_ui_test_generation\` with \`playwrightInput\` set to the absolute zip path from step 5 and \`modularizeCode: false\`.`;
     const modularizeNote = modularize
         ? `- **After generating the test**, run \`skyramp_modularization\` for code quality.`
-        : `- Do NOT run \`skyramp_modularization\` — skip modularization in CI.`;
+        : opts?.modularizeViaGenerationResult
+            ? `- Run \`skyramp_modularization\` ONLY when the generation result's CRITICAL NEXT STEPS instruct it (with the arguments they give) — never on your own initiative.`
+            : `- Do NOT run \`skyramp_modularization\` — skip modularization in CI.`;
     return `## Skyramp UI Test Recording
 
 ${getPersonaPrefix()} For UI recording, every action must be grounded in what \`browser_snapshot\` returns. If an element is not visible in the snapshot, do not interact with it.

package/build/prompts/code-reuse.d.ts

@@ -1,3 +1,2 @@
 import type { ScopedPomInput } from "../utils/pom-scope/index.js";
-export declare function isPomAwareTarget(language: string, framework?: string): boolean;
-export declare function getCodeReusePrompt(testFile: string, language: string, framework?: string, scopedPoms?: ScopedPomInput): string;
+export declare function getCodeReusePrompt(testFile: string, language: string, framework?: string, scopedPoms?: ScopedPomInput, testType?: string): string;