agy-context-saver 1.7.0 → 1.8.1

package/mcp/index.js

@@ -12,7 +12,6 @@
  * - Tools:
  *   - check_context_health: Inspects conversation transcripts using async streaming
  *     to diagnose turn count, tool polling loops, and context degradation with minimal memory.
- *   - subagent_brief: Formulates scope-isolated prompts for delegated subagents.
  *   - get_installation_status: Audits live Antigravity plugin, hook, MCP, and RTK integration.
  *   - sync_installation: Re-verifies and synchronizes installation and tool schemas.
  *   - read_transcript: Clean Markdown streaming reader for compact and full transcripts.
@@ -31,48 +30,57 @@ import { fileURLToPath } from "node:url";
 import { runInstall, runUninstall, runStatus, runRestoreBackups, detectExistingInstallation } from "../scripts/install-register.mjs";
 import { findExistingRtk } from "../scripts/provision-rtk.mjs";
 
+const __filename = fileURLToPath(import.meta.url);
+const realOrSelf = (p) => { try { return fs.realpathSync(p); } catch { return path.resolve(p); } };
+const isEntryPoint = Boolean(
+  process.argv[1] && realOrSelf(process.argv[1]) === realOrSelf(__filename)
+);
+
 // Handle CLI subcommands (e.g. npx agy-context-saver install)
-const cliArgs = process.argv.slice(2);
-const cliArg = cliArgs[0];
-const checkFlag = cliArgs.includes("--check") || cliArgs.includes("-c");
-const statusFlag = cliArgs.includes("--status") || cliArgs.includes("status");
-const uninstallFlag = cliArgs.includes("--uninstall") || cliArgs.includes("uninstall");
-const restoreFlag = cliArgs.includes("--restore-backups") || cliArgs.includes("restore-backups");
-const helpFlag = cliArgs.includes("--help") || cliArgs.includes("-h") || cliArgs.includes("help");
-const allowDegraded = cliArgs.includes("--allow-degraded");
-
-if (helpFlag) {
-  printHelp();
-  process.exit(0);
-}
-if ((checkFlag || statusFlag) && (uninstallFlag || restoreFlag)) {
-  console.error("Error: Read-only flags (--check, --status) cannot be combined with mutating flags (--uninstall, --restore-backups).");
-  process.exit(1);
-}
-if (checkFlag && statusFlag) {
-  console.error("Error: Conflicting flags: specify either --check or --status, not both.");
-  process.exit(1);
-}
-if (checkFlag) {
-  await runInstall({ checkOnly: true, allowDegraded });
-  process.exit(0);
-} else if (statusFlag) {
-  runStatus();
-  process.exit(0);
-} else if (uninstallFlag) {
-  await runUninstall({ restoreBackups: restoreFlag });
-  process.exit(0);
-} else if (restoreFlag) {
-  await runRestoreBackups();
-  process.exit(0);
-} else if (cliArg === "install" || cliArg === "--install") {
-  await runInstall({ allowDegraded });
-  process.exit(0);
+if (isEntryPoint) {
+  const cliArgs = process.argv.slice(2);
+  const cliArg = cliArgs[0];
+  const checkFlag = cliArgs.includes("--check") || cliArgs.includes("-c");
+  const statusFlag = cliArgs.includes("--status") || cliArgs.includes("status");
+  const uninstallFlag = cliArgs.includes("--uninstall") || cliArgs.includes("uninstall");
+  const restoreFlag = cliArgs.includes("--restore-backups") || cliArgs.includes("restore-backups");
+  const helpFlag = cliArgs.includes("--help") || cliArgs.includes("-h") || cliArgs.includes("help");
+  const allowDegraded = cliArgs.includes("--allow-degraded");
+
+  if (helpFlag) {
+    printHelp();
+    process.exit(0);
+  }
+  if ((checkFlag || statusFlag) && (uninstallFlag || restoreFlag)) {
+    console.error("Error: Read-only flags (--check, --status) cannot be combined with mutating flags (--uninstall, --restore-backups).");
+    process.exit(1);
+  }
+  if (checkFlag && statusFlag) {
+    console.error("Error: Conflicting flags: specify either --check or --status, not both.");
+    process.exit(1);
+  }
+  if (checkFlag) {
+    await runInstall({ checkOnly: true, allowDegraded });
+    process.exit(0);
+  } else if (statusFlag) {
+    runStatus();
+    process.exit(0);
+  } else if (uninstallFlag) {
+    await runUninstall({ restoreBackups: restoreFlag });
+    process.exit(0);
+  } else if (restoreFlag) {
+    await runRestoreBackups();
+    process.exit(0);
+  } else if (cliArg === "install" || cliArg === "--install") {
+    await runInstall({ allowDegraded });
+    process.exit(0);
+  }
 }
 
+
 function printHelp() {
   console.log(`
-Agy-Context-Saver 🛡️ (v1.7.0)
+Agy-Context-Saver 🛡️ (v1.8.1)
 Universal MCP Server & Lifecycle Governor for Google Antigravity (Powered by RTK)
 
 Usage:
@@ -99,19 +107,17 @@ MCP Server Stdio Mode:
 
 Testing & Verification Playbook:
   npm run status               Verify live 4-layer integration status
-  npm test                     Run complete 5-suite automated test suite
-  node tests/test-hook.mjs     Run 28 closed-topology governance rule tests
+  npm test                     Run complete 10-suite automated test suite
+  node tests/test-hook.mjs     Run 27 closed-topology governance rule tests
   node tests/test-mcp.mjs      Run MCP protocol, method, and schema tests
 
-Platform-Aware Inspection Directives (RTK):
-  Windows:
-    - Search:     rtk rg "<pattern>" .   (trailing dot is required to prevent stdin pipe stalls)
-    - Read:       rtk read <file>
-    - Directory:  rtk find <path>        (cross-platform; avoids missing ls.exe)
-  POSIX (macOS/Linux):
-    - Search:     rtk grep "<pattern>"  or  rtk rg "<pattern>"
-    - Read:       rtk read <file>
-    - Directory:  rtk ls <path>         or  rtk tree
+Unified 3-Tier Inspection Directives (Waymark + Hybrid view_file + RTK):
+  Tier 1 (Structure):  waymark_ask(..., plain: true) across all 23 supported formats
+  Tier 2 (Read/Edit):  view_file(StartLine, EndLine) (1-150L default; or explicit 1..TotalLines full-file audit <= 800L & <= 45 KiB)
+  Tier 3 (RTK CLI):
+    - Spillovers/Logs: rtk read <path> (--head-lines / --tail-lines)
+    - Directory:       rtk find <path> (Windows & POSIX) or rtk ls <path> (POSIX)
+    - Prose Search:    rtk rg "<pattern>" . (Windows trailing dot required) or rtk grep "<pattern>" (POSIX)
 `);
 }
 
@@ -119,7 +125,7 @@ const __dirname = path.dirname(fileURLToPath(import.meta.url));
 const rulesPath = path.resolve(__dirname, "../rules/AGENTS.md");
 
 const SERVER_NAME = "agy-context-saver";
-const SERVER_VERSION = "1.7.0";
+const SERVER_VERSION = "1.8.1";
 
 // --- Tools Specification ---
 const TOOLS = [
@@ -140,68 +146,6 @@ const TOOLS = [
       }
     }
   },
-  {
-    name: "subagent_brief",
-    description: "Generates scope-isolated, anti-bloat prompts and ready-to-use invoke_subagent JSON payloads for single or multiple delegated subagents. Instructs subagents to absorb intermediate file reads/searches and return only high-signal findings.",
-    inputSchema: {
-      type: "object",
-      properties: {
-        objective: {
-          type: "string",
-          description: "The core research or investigative objective (for single subagent)."
-        },
-        role: {
-          type: "string",
-          description: "2-5 word role description (e.g. 'Comprehensive Test Engineer', 'Codebase Security Auditor'). Defaults to 'Scope-Isolated Deep Research Subagent'."
-        },
-        instructions: {
-          type: "string",
-          description: "Optional step-by-step instructions or specialized directives."
-        },
-        scopeFiles: {
-          type: "array",
-          items: { type: "string" },
-          description: "List of files or directories to investigate."
-        },
-        expectedDeliverable: {
-          type: "string",
-          description: "The concise format expected back by the parent agent."
-        },
-        typeName: {
-          type: "string",
-          description: "Subagent type name: 'self' (default, inherits parent tools/prompt/model) or 'research' (read-only)."
-        },
-        model: {
-          type: "string",
-          enum: ["inherit", "flash_lite", "flash", "pro"],
-          description: "Model tier: 'inherit' (default), 'flash_lite', 'flash', or 'pro'."
-        },
-        workspace: {
-          type: "string",
-          enum: ["inherit", "branch", "share"],
-          description: "Workspace isolation: 'inherit' (default), 'branch', or 'share'."
-        },
-        subagents: {
-          type: "array",
-          description: "Optional array of subagent configurations to formulate multiple subagents at once.",
-          items: {
-            type: "object",
-            properties: {
-              role: { type: "string", description: "2-5 word role description." },
-              objective: { type: "string", description: "The core objective for this subagent." },
-              instructions: { type: "string", description: "Optional step-by-step instructions." },
-              scopeFiles: { type: "array", items: { type: "string" }, description: "Files/directories to investigate." },
-              expectedDeliverable: { type: "string", description: "Expected deliverable format." },
-              typeName: { type: "string", description: "Subagent type name: 'self' or 'research'." },
-              model: { type: "string", enum: ["inherit", "flash_lite", "flash", "pro"], description: "Model tier." },
-              workspace: { type: "string", enum: ["inherit", "branch", "share"], description: "Workspace isolation." }
-            },
-            required: ["objective"]
-          }
-        }
-      }
-    }
-  },
   {
     name: "get_installation_status",
     description: "Inspect the live installation status of Agy-Context-Saver across all Antigravity integration points (Native Plugin Link, Governor Hook in hooks.json, MCP Server in mcp_config.json, and Tool Schemas).",
@@ -254,7 +198,7 @@ const TOOLS = [
   },
   {
     name: "query_transcript",
-    description: "Granular query and forensic filtering engine for Antigravity conversation transcripts. Searches by keyword or regex, filters by role (user, assistant, tool, error), slices step ranges, and automatically dereferences full content when truncated.",
+    description: "Granular query and forensic filtering engine for Antigravity conversation transcripts. Searches by keyword or regex, filters by role (user, assistant, subagent, tool, error), slices step ranges, and automatically dereferences full content when truncated.",
     inputSchema: {
       type: "object",
       properties: {
@@ -269,7 +213,7 @@ const TOOLS = [
         roles: {
           type: "array",
           items: { type: "string" },
-          description: "Filter steps by role/source: 'user', 'assistant', 'tool', 'error', or 'all' (default: ['user', 'assistant'])."
+          description: "Filter steps by role/source: 'user', 'assistant', 'subagent', 'tool', 'error', or 'all' (default: ['user', 'assistant'])."
         },
         mode: {
           type: "string",
@@ -323,7 +267,7 @@ const RESOURCES = [
   {
     uri: "context-saver://rules/governance",
     name: "Antigravity Execution & Context Governance Rulebook",
-    description: "Authoritative 3-Layer rules preventing background polling loops, maximizing synchronous execution, and isolating context via subagents.",
+    description: "Authoritative 3-Layer rules preventing background polling loops, maximizing synchronous execution, and keeping command output compact.",
     mimeType: "text/markdown"
   }
 ];
@@ -449,7 +393,7 @@ async function handleCheckContextHealth(args = {}) {
       reportLines.push("⚠️ Recommendation: Background polling detected! Cease manage_task(status) calls and yield execution to native Reactive Wakeup.");
     }
     if (userTurns >= 35) {
-      reportLines.push("💡 Recommendation: Turn budget approaching threshold. Delegate broad research to subagents to preserve context.");
+      reportLines.push("💡 Recommendation: Turn budget approaching threshold. Summarize progress and continue in a fresh conversation to preserve context.");
     } else {
       reportLines.push("✓ Context footprint is well-managed.");
     }
@@ -463,108 +407,10 @@ async function handleCheckContextHealth(args = {}) {
   }
 }
 
-function formatSingleBrief({
-  role = "Scope-Isolated Deep Research Subagent",
-  objective = "",
-  instructions = "",
-  scopeFiles = [],
-  expectedDeliverable = "Synthesized factual findings"
-} = {}) {
-  const scopeStr = scopeFiles && scopeFiles.length ? `\n\nTarget files:\n${scopeFiles.map(f => `- ${f}`).join("\n")}` : "";
-  const instrStr = instructions && instructions.trim() ? `\n\nSPECIFIC INSTRUCTIONS:\n${instructions.trim()}` : "";
-  const hasBrainUuidPath = /[/\\]\.gemini[/\\]antigravity[/\\]brain[/\\][0-9a-f-]{36}/i.test(`${objective}\n${instructions}\n${expectedDeliverable}`);
-  const brainPathWarning = hasBrainUuidPath
-    ? `\n\n⚠️ ARTIFACT DIRECTORY WRITE GUARD: Any 'brain/<uuid>/' path referenced above belongs to another conversation. You may READ from it via 'view_file', but when calling 'write_to_file' or 'replace_file_content', you MUST either write to workspace '<workspace>/scratch/...' (OMIT ArtifactMetadata) or substitute YOUR OWN Artifact Directory Path from <user_information> (INCLUDE ArtifactMetadata).`
-    : "";
-  return [
-    `ROLE: ${role}`,
-    `OBJECTIVE: ${objective}${scopeStr}${instrStr}${brainPathWarning}`,
-    ``,
-    `CRITICAL CONTEXT GOVERNANCE INSTRUCTIONS (POINTER-OVER-WIRE):`,
-    `1. Isolated Workspace & Bounded File Slices: Do all multi-file exploration in this subagent sandbox. Prefer 'view_file' with bounded slices (StartLine and EndLine, max 150 lines) over 'rtk read' so line numbers are preserved and terminal buffers never truncate file headers.`,
-    `2. Pointer-Over-Wire (POW) Protocol: In-band messaging channels (send_message) have strict transport payload limits (~16-20 KB). If your findings or deliverable exceed 1 KB:`,
-    `   a. Write the full comprehensive deliverable using 'write_to_file' following Antigravity's strict path rules:`,
-    `      - Option A (Workspace 'scratch/' — Recommended): Set TargetFile to '<workspace>/scratch/<task>_findings.md' and OMIT ArtifactMetadata completely (passing ArtifactMetadata on a workspace path throws "files must be written to the correct artifact directory").`,
-    `      - Option B (Your Own Brain 'scratch/'): Set TargetFile to '<your-Artifact-Directory-Path>/scratch/<task>_findings.md' using the exact Artifact Directory Path from YOUR <user_information> block (NEVER a parent's brain/<other-uuid>/ path) and INCLUDE ArtifactMetadata: { Summary: "...", UserFacing: false, RequestFeedback: false }.`,
-    `   b. In 'send_message', transmit ONLY:`,
-    `      - Executive Summary (3-5 high-signal sentences)`,
-    `      - Verdict / Scoreboard Table`,
-    `      - Clickable File Link: [Full Deliverable](file:///absolute/path/to/findings.md)`,
-    `3. Fast Turn Termination: Once 'send_message' is dispatched, stop calling tools and conclude your turn immediately so the parent agent wakes up without latency.`,
-    `4. Format: ${expectedDeliverable}`,
-    `5. Citation Mandate: Keep your final response concise, structured, and cite exact file:line references.`
-  ].join("\n");
-}
-
-function handleSubagentBrief(args = {}) {
-  let specs = [];
-  if (Array.isArray(args.subagents) && args.subagents.length > 0) {
-    specs = args.subagents;
-  } else if (args.objective || args.role || args.instructions) {
-    specs = [args];
-  } else {
-    return {
-      isError: true,
-      content: [{ type: "text", text: "subagent_brief error: provide either 'objective' (for a single subagent) or 'subagents' array (for multiple subagents)." }]
-    };
-  }
-
-  const subagentsPayload = [];
-  const renderedBriefs = [];
-
-  for (let i = 0; i < specs.length; i++) {
-    const spec = specs[i] || {};
-    const role = spec.role || spec.Role || "Scope-Isolated Deep Research Subagent";
-    const objective = spec.objective || spec.Objective || "";
-    const instructions = spec.instructions || spec.Instructions || "";
-    const scopeFiles = spec.scopeFiles || spec.ScopeFiles || [];
-    const expectedDeliverable = spec.expectedDeliverable || spec.ExpectedDeliverable || "Synthesized factual findings";
-    const typeName = spec.typeName || spec.TypeName || "self";
-    const model = spec.model || spec.Model || "inherit";
-    const workspace = spec.workspace || spec.Workspace || "inherit";
-
-    const promptText = formatSingleBrief({
-      role,
-      objective,
-      instructions,
-      scopeFiles,
-      expectedDeliverable
-    });
-
-    subagentsPayload.push({
-      TypeName: typeName,
-      Role: role,
-      Model: model,
-      Workspace: workspace,
-      Prompt: promptText
-    });
-
-    if (specs.length === 1) {
-      renderedBriefs.push(promptText);
-    } else {
-      renderedBriefs.push(`### Subagent ${i + 1}: ${role}\n${promptText}`);
-    }
-  }
-
-  const payloadJson = JSON.stringify({ Subagents: subagentsPayload }, null, 2);
-
-  const fullText = [
-    renderedBriefs.join("\n\n---\n\n"),
-    "",
-    "==================================================",
-    "READY-TO-USE invoke_subagent PAYLOAD:",
-    "```json",
-    payloadJson,
-    "```"
-  ].join("\n");
-
-  return { content: [{ type: "text", text: fullText }] };
-}
-
 function handleGetInstallationStatus() {
   const existing = detectExistingInstallation();
   const rtkInfo = findExistingRtk();
-  const schemaText = existing.details.schemas.exists ? "ALL 6 SCHEMAS PRESENT" : "MISSING";
+  const schemaText = existing.details.schemas.exists ? "ALL 5 SCHEMAS PRESENT" : "MISSING";
   const report = [
     `### Agy-Context-Saver Installation Status: ${existing.isComplete ? "HEALTHY & ACTIVE 🛡️" : existing.isInstalled ? "PARTIAL INSTALLATION ⚠️" : "NOT INSTALLED ❌"}`,
     `- Native Plugin Link: ${existing.details.plugin.exists ? `ACTIVE (${existing.details.plugin.target})` : "NOT LINKED"}`,
@@ -594,7 +440,7 @@ async function handleSyncInstallation({ checkOnly = false } = {}) {
     `- Plugin Link: ${result.details?.plugin?.exists ? "Verified" : "Updated"}`,
     `- Lifecycle Hook: Registered in hooks.json`,
     `- MCP Server: Registered in mcp_config.json`,
-    `- Tool Schemas: 6 Schemas mirrored to ~/.gemini/antigravity/mcp/agy-context-saver`,
+    `- Tool Schemas: 5 Schemas mirrored to ~/.gemini/antigravity/mcp/agy-context-saver`,
     `- RTK Binary: ${rtkInfo ? `Verified (${rtkInfo.version})` : "Provisioned"}`,
     "",
     checkOnly ? "✓ Pre-flight check complete (dry-run)." : "✓ Installation fully synchronized and up-to-date in Zero-Delay mode."
@@ -740,6 +586,11 @@ function cleanMessageContent(content, maxChars = MAX_TURN_CHARS) {
     text = text.replace(/[A-Za-z0-9+/=]{200,}/g, "[Binary/Base64 Payload Omitted]");
   }
 
+  // Strip Git CRLF/LF line-ending diagnostic warning noise (wastes context window tokens)
+  if (text.includes("will be replaced by")) {
+    text = text.replace(/(?:#\s*)?warning:\s*(?:in the working copy of [^,\r\n]+,\s*)?(?:CRLF will be replaced by LF|LF will be replaced by CRLF)[^\r\n]*(?:\r?\n)?/gi, "");
+  }
+
   // Clamp lines exceeding MAX_LINE_CHARS only if newline exists and text is long enough
   if (text.length > MAX_LINE_CHARS && text.includes("\n")) {
     const lines = text.split("\n");
@@ -958,13 +809,20 @@ async function handleReadTranscript({ conversationId, mode = "compact", lastTurn
   const formattedBlocks = [];
   let currentLength = header.length;
   let clamped = false;
+  const maxWindowChars = mode === "compact" ? 3800 : MAX_OUTPUT_CHARS;
 
   for (let i = selected.length - 1; i >= 0; i--) {
-    const block = formatTranscriptItem(selected[i], {
+    const item = selected[i];
+    const isDialogueTurn =
+      item.type === "USER_INPUT" ||
+      item.type === "PLANNER_RESPONSE" ||
+      item.type === "SUBAGENT_RESPONSE";
+    const block = formatTranscriptItem(item, {
       includeThinking,
-      includeToolCalls: true
+      includeToolCalls: true,
+      maxContentChars: mode === "compact" && !isDialogueTurn ? 300 : MAX_TURN_CHARS
     });
-    if (currentLength + block.length + 6 > MAX_OUTPUT_CHARS) {
+    if (currentLength + block.length + 6 > maxWindowChars && formattedBlocks.length > 0) {
       clamped = true;
       break;
     }
@@ -974,7 +832,8 @@ async function handleReadTranscript({ conversationId, mode = "compact", lastTurn
 
   let fullOutput = header + formattedBlocks.join("\n\n---\n\n");
   if (clamped) {
-    fullOutput += "\n\n... [Transcript display clamped at 24 KB safety ceiling] ...";
+    const ceilingLabel = mode === "compact" ? "3.8 KB inline" : "24 KB";
+    fullOutput += `\n\n... [Transcript display clamped at ${ceilingLabel} safety ceiling; use query_transcript(stepIndex: N) for specific steps] ...`;
   }
 
   return { content: [{ type: "text", text: fullOutput }] };
@@ -1325,8 +1184,6 @@ async function handleRequest(request) {
 
     if (name === "check_context_health") {
       toolResult = await handleCheckContextHealth(args);
-    } else if (name === "subagent_brief") {
-      toolResult = handleSubagentBrief(args);
     } else if (name === "get_installation_status") {
       toolResult = handleGetInstallationStatus();
     } else if (name === "sync_installation") {
@@ -1414,33 +1271,62 @@ async function handleRequest(request) {
   };
 }
 
-// Stdio JSON-RPC line loop
-const MAX_STDIO_PAYLOAD_BYTES = Number(process.env.AGY_MAX_STDIO_PAYLOAD_BYTES) || (10 * 1024 * 1024); // 10 MB ceiling default
-const rl = readline.createInterface({ input: process.stdin, output: process.stdout, terminal: false });
+if (isEntryPoint) {
+  // Stdio JSON-RPC line loop
+  const MAX_STDIO_PAYLOAD_BYTES = Number(process.env.AGY_MAX_STDIO_PAYLOAD_BYTES) || (10 * 1024 * 1024); // 10 MB ceiling default
+  const rl = readline.createInterface({ input: process.stdin, output: process.stdout, terminal: false });
 
-rl.on("line", async (line) => {
-  if (!line || !line.trim()) return;
-  if (line.length > MAX_STDIO_PAYLOAD_BYTES) {
-    const errResp = {
-      jsonrpc: "2.0",
-      id: null,
-      error: { code: -32600, message: "Invalid Request: Payload exceeds 10MB safety ceiling" }
-    };
-    process.stdout.write(JSON.stringify(errResp) + "\n");
-    return;
-  }
-  try {
-    const req = JSON.parse(line);
-    const resp = await handleRequest(req);
-    if (resp) {
-      process.stdout.write(JSON.stringify(resp) + "\n");
+  rl.on("line", async (line) => {
+    if (!line || !line.trim()) return;
+    if (line.length > MAX_STDIO_PAYLOAD_BYTES) {
+      const errResp = {
+        jsonrpc: "2.0",
+        id: null,
+        error: { code: -32600, message: "Invalid Request: Payload exceeds 10MB safety ceiling" }
+      };
+      process.stdout.write(JSON.stringify(errResp) + "\n");
+      return;
     }
-  } catch (err) {
-    const errResp = {
-      jsonrpc: "2.0",
-      id: null,
-      error: { code: -32700, message: `Parse error: ${err.message}` }
-    };
-    process.stdout.write(JSON.stringify(errResp) + "\n");
+    try {
+      const req = JSON.parse(line);
+      const resp = await handleRequest(req);
+      if (resp) {
+        process.stdout.write(JSON.stringify(resp) + "\n");
+      }
+    } catch (err) {
+      const errResp = {
+        jsonrpc: "2.0",
+        id: null,
+        error: { code: -32700, message: `Parse error: ${err.message}` }
+      };
+      process.stdout.write(JSON.stringify(errResp) + "\n");
+    }
+  });
+
+  // Anti-Orphan Lifecycle Governance: Prevent lingering Node processes when parent IDE closes
+  rl.on("close", () => process.exit(0));
+  process.stdin.on("end", () => process.exit(0));
+  process.stdin.on("close", () => process.exit(0));
+  process.stdout.on("error", (err) => {
+    if (err && err.code === "EPIPE") {
+      process.exit(0);
+    }
+  });
+
+  // Lightweight parent-process watchdog: If parent IDE process terminates, exit child
+  const initialPpid = process.ppid;
+  if (initialPpid && initialPpid > 1) {
+    const watchdog = setInterval(() => {
+      try {
+        process.kill(initialPpid, 0);
+      } catch (err) {
+        if (err && err.code === "ESRCH") {
+          process.exit(0);
+        }
+      }
+    }, 15000);
+    watchdog.unref();
   }
-});
+}
+
+export { handleRequest };

package/mcp/instructions.md

@@ -11,16 +11,18 @@ Agy-Context-Saver is the universal Model Context Protocol server and lifecycle g
 2. **Never Read Internal State Directly**:
    - Access to `.system_generated/` via `view_file` is strictly denied to prevent transcript explosion.
    - Use `read_transcript` or `query_transcript` to inspect conversation history cleanly.
-3. **Platform-Aware Codebase Inspection**:
-   - **File Reading**: Prefer `view_file` with bounded slices (`StartLine` and `EndLine`, max 150 lines) to preserve line numbers and avoid terminal truncation.
-   - **Windows Search/Listing**: Use `rtk find <path>` and `rtk rg "<pattern>" .` (the trailing `.` is required to avoid stdin stalls).
-   - **POSIX Search/Listing**: Use `rtk ls <path>` and `rtk grep "<pattern>"` or `rtk rg "<pattern>"`.
+3. **Unified 3-Tier Codebase & File Inspection**:
+   - **Tier 1 — Structural Discovery (`waymark_ask` with `plain: true`)**: First locate exact `[StartLine, EndLine]` coordinates via `waymark_ask(..., plain: true)` across all 23 supported formats (code ASTs + `markdown`, `yaml`, `json`, `toml`, `bash`). Always set `plain: true` (~19–466 tokens, zero shell quoting or stdin hazards).
+   - **Tier 2 — Precision Workspace File Reading (`view_file` Hybrid Slice Policy)**: Pass the `StartLine` and `EndLine` span from `waymark_ask` into `view_file` (bounded slices of `1–150` lines) to preserve 1-indexed line coordinates for `replace_file_content` and avoid terminal stdout truncation. Partial slices of `151–800` lines are denied; an explicit one-pass full-file audit (`StartLine: 1, EndLine: <TotalLines>`) is permitted only when the file on disk is $\le 800$ lines and $\le 45\text{ KiB}$ (`46,080` bytes).
+   - **Tier 3 — RTK Boundaries (`rtk read`, `rtk find`, `rtk rg`)**:
+     - **Spillovers & Task Logs**: Read `.system_generated/steps/<step>/output.txt` spillovers and completed `tasks/<taskId>.log` files via `rtk read <path>` (with `--head-lines` / `--tail-lines`). Never use `rtk read -l minimal|aggressive` or `-m` to derive line numbers for edits.
+     - **Directory/File Discovery**: Use `rtk find <path>` (Windows & POSIX) or `rtk ls <path>` (POSIX).
+     - **Prose / Regex Search**: Use `rtk rg "<pattern>" <path>` (explicit `<path>` required on Windows to avoid stdin stalls) strictly for free-text/regex searches inside body prose or string values when the symbol, heading, or config key name is unknown.
 
 ## Available MCP Tools
 - `check_context_health`: Diagnoses transcript turn counts, byte payload, and polling loops. Accepts `transcriptPath` or `conversationId`.
-- `subagent_brief`: Generates scope-isolated instructions for delegated subagents to prevent parent context bloat.
 - `read_transcript`: Streams recent turns from `transcript.jsonl` or `transcript_full.jsonl` formatted as clean Markdown.
-- `query_transcript`: Forensic regex search and filtering engine across transcript steps with optional `summaryOnly: true`.
+- `query_transcript`: Forensic regex search and filtering engine across transcript steps with optional `summaryOnly: true`, plus surgical step extraction via `stepIndex` and `field`.
 - `get_installation_status`: Audits live health of the plugin link, hook, MCP server, tool schemas, and RTK.
 - `sync_installation`: Synchronizes and repairs all 4 integration layers in ~25ms.
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "agy-context-saver",
-  "version": "1.7.0",
+  "version": "1.8.1",
   "description": "Universal Model Context Protocol (MCP) server & lifecycle governor purpose-built for Google Antigravity",
   "main": "mcp/index.js",
   "bin": {
@@ -13,8 +13,7 @@
     "uninstall": "node scripts/install-register.mjs --uninstall",
     "status": "node scripts/install-register.mjs --status",
     "preflight": "node scripts/release-preflight.mjs",
-    "test:synergy": "node tests/test-waymark-synergy.mjs",
-    "test": "node tests/test-hook.mjs && node tests/test-mcp.mjs && node tests/test-output-governance.mjs && node tests/test-e2e-probes.mjs && node tests/test-performance-probes.mjs && node tests/test-installed-verification.mjs && node tests/test-lifecycle-rollback.mjs && node tests/test-subagent-trials.mjs && node tests/test-check-only-readonly.mjs && node tests/test-malformed-config-preservation.mjs && node tests/test-waymark-synergy.mjs"
+    "test": "node tests/test-hook.mjs && node tests/test-mcp.mjs && node tests/test-output-governance.mjs && node tests/test-e2e-probes.mjs && node tests/test-performance-probes.mjs && node tests/test-installed-verification.mjs && node tests/test-lifecycle-rollback.mjs && node tests/test-check-only-readonly.mjs && node tests/test-malformed-config-preservation.mjs && node tests/test-rtk-ledger-fresh-install.mjs"
   },
   "repository": {
     "type": "git",

package/plugin.json

@@ -1,12 +1,12 @@
 {
   "name": "agy-context-saver",
   "displayName": "Antigravity Context Shield & Governor",
-  "version": "1.7.0",
-  "description": "Deterministic background polling governor, RTK terminal optimizer, and subagent context preservation shield for Antigravity.",
+  "version": "1.8.1",
+  "description": "Deterministic background polling governor, RTK terminal optimizer, and transcript context shield for Antigravity.",
   "logo": "assets/logo.svg",
   "suggestedPrompts": [
     "Audit context health and token consumption for this session",
-    "Formulate an isolated subagent brief under Pointer-Over-Wire protocol",
+    "Search this conversation's transcript for earlier tool errors",
     "Inspect recent conversation turns without raw JSON transcript bloat"
   ]
 }