@levnikolaevich/hex-line-mcp 1.3.1 → 1.3.3

package/benchmark.mjs

@@ -1,1106 +0,0 @@
-#!/usr/bin/env node
-/**
- * Hex-line Combo Benchmark v3
- *
- * Compares "agent without hex-line" vs "agent with hex-line" across
- * read-only and write scenarios. Measures chars in response (proxy for tokens).
- *
- * Usage: node mcp/hex-line-mcp/benchmark.mjs [--repo /path/to/repo] [--with-graph]
- * Default repo: current working directory.
- *
- * Zero external deps beyond hex-line lib modules.
- */
-
-import { readFileSync, writeFileSync, unlinkSync, readdirSync, mkdirSync, rmSync } from "node:fs";
-import { performance } from "node:perf_hooks";
-import { resolve, basename } from "node:path";
-import { tmpdir } from "node:os";
-import { fnv1a, lineTag, rangeChecksum } from "./lib/hash.mjs";
-import { readFile } from "./lib/read.mjs";
-import { directoryTree } from "./lib/tree.mjs";
-import { fileInfo } from "./lib/info.mjs";
-import { verifyChecksums } from "./lib/verify.mjs";
-import { fileChanges } from "./lib/changes.mjs";
-import { editFile } from "./lib/edit.mjs";
-import { grepSearch } from "./lib/search.mjs";
-import { bulkReplace } from "./lib/bulk-replace.mjs";
-import { fileOutline } from "./lib/outline.mjs";
-import {
-    walkDir, getFileLines, categorize, generateTempCode,
-    simBuiltInReadFull, simBuiltInOutlineFull, simBuiltInGrep,
-    simBuiltInLsR, simBuiltInStat, simBuiltInWrite, simBuiltInEdit, simBuiltInVerify,
-    simHexLineOutlinePlusRead, simHexLineGrep, simHexLineWrite, simHexLineEditDiff,
-    runN, fmt, pctSavings, RUNS,
-} from "./lib/benchmark-helpers.mjs";
-// ---------------------------------------------------------------------------
-// CLI
-// ---------------------------------------------------------------------------
-
-const args = process.argv.slice(2);
-let repoRoot = process.cwd();
-const repoIdx = args.indexOf("--repo");
-if (repoIdx !== -1 && args[repoIdx + 1]) {
-    repoRoot = resolve(args[repoIdx + 1]);
-}
-
-const withGraph = args.includes("--with-graph");
-
-// ---------------------------------------------------------------------------
-// Main
-// ---------------------------------------------------------------------------
-
-async function main() {
-    const allFiles = walkDir(repoRoot);
-    if (allFiles.length === 0) {
-        console.log(`No code files found in ${repoRoot}`);
-        process.exit(1);
-    }
-
-    const totalLines = allFiles.reduce((sum, f) => {
-        const lines = getFileLines(f);
-        return lines ? sum + lines.length : sum;
-    }, 0);
-
-    const cats = categorize(allFiles);
-    const repoName = basename(repoRoot);
-
-    // Top 3 largest code files for realistic tests
-    const sorted = allFiles.map(f => ({ f, lines: getFileLines(f)?.length || 0 }))
-        .sort((a, b) => b.lines - a.lines);
-    const largeFiles = sorted.slice(0, 3).map(s => s.f);
-
-    // Temp file setup
-    const ts = Date.now();
-    const tmpPath = resolve(tmpdir(), `hex-line-bench-${ts}.js`);
-    const tmpLines = generateTempCode();
-    const tmpContent = tmpLines.join("\n");
-    writeFileSync(tmpPath, tmpContent, "utf-8");
-
-    const results = [];
-
-    // ===================================================================
-    // TEST 1: Read full file
-    // ===================================================================
-    for (const [cat, files] of Object.entries(cats)) {
-        if (files.length === 0) continue;
-        const withoutArr = [];
-        const withArr = [];
-
-        for (const f of files) {
-            const lines = getFileLines(f);
-            if (!lines) continue;
-            withoutArr.push(runN(() => simBuiltInReadFull(f, lines).length));
-            withArr.push(runN(() => readFile(f).length));
-        }
-
-        if (withoutArr.length === 0) continue;
-        const avgWithout = Math.round(withoutArr.reduce((a, b) => a + b.value, 0) / withoutArr.length);
-        const avgWith = Math.round(withArr.reduce((a, b) => a + b.value, 0) / withArr.length);
-        const avgMsWithout = parseFloat((withoutArr.reduce((a, b) => a + b.ms, 0) / withoutArr.length).toFixed(1));
-        const avgMsWith = parseFloat((withArr.reduce((a, b) => a + b.ms, 0) / withArr.length).toFixed(1));
-
-        const label = { small: "<50L", medium: "50-200L", large: "200-500L", xl: "500L+" }[cat];
-        results.push({
-            num: 1, scenario: `Read full (${label})`,
-            without: avgWithout, withSL: avgWith,
-            savings: pctSavings(avgWithout, avgWith),
-            latencyWithout: avgMsWithout, latencyWith: avgMsWith,
-        });
-    }
-
-    // ===================================================================
-    // TEST 2: Read with outline — full read vs outline + targeted read
-    // ===================================================================
-    for (const cat of ["large", "xl"]) {
-        const files = cats[cat] || [];
-        if (files.length === 0) continue;
-        const withoutArr = [];
-        const withArr = [];
-
-        for (const f of files) {
-            const lines = getFileLines(f);
-            if (!lines) continue;
-            withoutArr.push(runN(() => simBuiltInOutlineFull(f, lines).length));
-            withArr.push(runN(() => simHexLineOutlinePlusRead(f, lines).length));
-        }
-
-        if (withoutArr.length === 0) continue;
-        const avgWithout = Math.round(withoutArr.reduce((a, b) => a + b.value, 0) / withoutArr.length);
-        const avgWith = Math.round(withArr.reduce((a, b) => a + b.value, 0) / withArr.length);
-        const avgMsWithout = parseFloat((withoutArr.reduce((a, b) => a + b.ms, 0) / withoutArr.length).toFixed(1));
-        const avgMsWith = parseFloat((withArr.reduce((a, b) => a + b.ms, 0) / withArr.length).toFixed(1));
-
-        const label = cat === "large" ? "200-500L" : "500L+";
-        results.push({
-            num: 2, scenario: `Outline+read (${label})`,
-            without: avgWithout, withSL: avgWith,
-            savings: pctSavings(avgWithout, avgWith),
-            latencyWithout: avgMsWithout, latencyWith: avgMsWith,
-        });
-    }
-
-    // ===================================================================
-    // TEST 3: Grep search
-    // ===================================================================
-    {
-        const grepFiles = [...(cats.medium || []), ...(cats.large || []), ...(cats.xl || [])].slice(0, 3);
-        if (grepFiles.length > 0) {
-            const withoutArr = [];
-            const withArr = [];
-
-            for (const f of grepFiles) {
-                const lines = getFileLines(f);
-                if (!lines) continue;
-                const pattern = "function|class|const";
-                withoutArr.push(runN(() => simBuiltInGrep(pattern, f).length));
-                withArr.push(runN(() => simHexLineGrep(f, lines, pattern).length));
-            }
-
-            if (withoutArr.length > 0) {
-                const avgWithout = Math.round(withoutArr.reduce((a, b) => a + b.value, 0) / withoutArr.length);
-                const avgWith = Math.round(withArr.reduce((a, b) => a + b.value, 0) / withArr.length);
-                const avgMsWithout = parseFloat((withoutArr.reduce((a, b) => a + b.ms, 0) / withoutArr.length).toFixed(1));
-                const avgMsWith = parseFloat((withArr.reduce((a, b) => a + b.ms, 0) / withArr.length).toFixed(1));
-                results.push({
-                    num: 3, scenario: "Grep search",
-                    without: avgWithout, withSL: avgWith,
-                    savings: pctSavings(avgWithout, avgWith),
-                    latencyWithout: avgMsWithout, latencyWith: avgMsWith,
-                });
-            }
-        }
-    }
-
-    // ===================================================================
-    // TEST 4: Directory tree
-    // ===================================================================
-    {
-        const { value: without, ms: withoutMs } = runN(() => simBuiltInLsR(repoRoot, 0, 3).length);
-        const { value: withSL, ms: withMs } = runN(() => directoryTree(repoRoot, { max_depth: 3 }).length);
-        results.push({
-            num: 4, scenario: "Directory tree",
-            without, withSL,
-            savings: pctSavings(without, withSL),
-            latencyWithout: withoutMs, latencyWith: withMs,
-        });
-    }
-
-    // ===================================================================
-    // TEST 5: File info
-    // ===================================================================
-    {
-        const infoFile = allFiles[Math.floor(allFiles.length / 2)] || allFiles[0];
-        const { value: without, ms: withoutMs } = runN(() => simBuiltInStat(infoFile).length);
-        const { value: withSL, ms: withMs } = runN(() => fileInfo(infoFile).length);
-        results.push({
-            num: 5, scenario: "File info",
-            without, withSL,
-            savings: pctSavings(without, withSL),
-            latencyWithout: withoutMs, latencyWith: withMs,
-        });
-    }
-
-    // ===================================================================
-    // TEST 6: Create file (write)
-    // ===================================================================
-    {
-        const { value: without, ms: withoutMs } = runN(() => simBuiltInWrite(tmpPath, tmpContent).length);
-        const { value: withSL, ms: withMs } = runN(() => simHexLineWrite(tmpPath, tmpContent).length);
-        results.push({
-            num: 6, scenario: "Create file (200L)",
-            without, withSL,
-            savings: pctSavings(without, withSL),
-            latencyWithout: withoutMs, latencyWith: withMs,
-        });
-    }
-
-    // ===================================================================
-    // TEST 7: Edit x5 sequential
-    // ===================================================================
-    {
-        const editTargets = [
-            { line: 13, new: '        this.configPath = resolve(configPath || ".");' },
-            { line: 55, new: "    const { retries = MAX_RETRIES, delay = 200, backoff = 3 } = options;" },
-            { line: 75, new: "        this.timeout = options.timeout ?? DEFAULT_TIMEOUT;" },
-            { line: 116, new: "        return this; // chainable" },
-            { line: 148, new: "    /** @type {string[]} */\n    const errors = [];" },
-        ];
-
-        let totalWithout = 0;
-        let totalWith = 0;
-        let totalMsWithout = 0;
-        let totalMsWith = 0;
-
-        for (const edit of editTargets) {
-            const origLines = [...tmpLines];
-            const newLines = [...tmpLines];
-            const idx = edit.line - 1;
-            if (idx < newLines.length) {
-                newLines[idx] = edit.new;
-            }
-
-            const rW = runN(() => simBuiltInEdit(tmpPath, origLines, newLines).length);
-            const rH = runN(() => simHexLineEditDiff(origLines, newLines).length);
-            totalWithout += rW.value;
-            totalWith += rH.value;
-            totalMsWithout += rW.ms;
-            totalMsWith += rH.ms;
-        }
-
-        results.push({
-            num: 7, scenario: "Edit x5 sequential",
-            without: totalWithout, withSL: totalWith,
-            savings: pctSavings(totalWithout, totalWith),
-            latencyWithout: parseFloat(totalMsWithout.toFixed(1)), latencyWith: parseFloat(totalMsWith.toFixed(1)),
-        });
-    }
-
-    // ===================================================================
-    // TEST 8: Verify checksums
-    // ===================================================================
-    {
-        const fileLines = readFileSync(tmpPath, "utf-8").replace(/\r\n/g, "\n").split("\n");
-        const hashes = fileLines.map(l => fnv1a(l));
-        const cs1 = rangeChecksum(hashes.slice(0, 50), 1, 50);
-        const cs2 = rangeChecksum(hashes.slice(50, 100), 51, 100);
-        const cs3 = rangeChecksum(hashes.slice(100, 150), 101, 150);
-        const cs4 = rangeChecksum(hashes.slice(150, 200), 151, 200);
-        const checksums = [cs1, cs2, cs3, cs4];
-
-        const { value: without, ms: withoutMs } = runN(() => simBuiltInVerify(tmpPath, fileLines).length);
-        const { value: withSL, ms: withMs } = runN(() => verifyChecksums(tmpPath, checksums).length);
-
-        results.push({
-            num: 8, scenario: "Verify checksums (4 ranges)",
-            without, withSL,
-            savings: pctSavings(without, withSL),
-            latencyWithout: withoutMs, latencyWith: withMs,
-        });
-    }
-
-    // ===================================================================
-    // TEST 9: Multi-file read (batch)
-    // ===================================================================
-    {
-        const batchFiles = (cats.small || []).slice(0, 3);
-        if (batchFiles.length >= 2) {
-            // Without hex-line: N separate Read calls
-            const { value: without, ms: withoutMs } = runN(() => {
-                let total = 0;
-                for (const f of batchFiles) {
-                    const lines = getFileLines(f);
-                    if (lines) total += simBuiltInReadFull(f, lines).length;
-                }
-                return total;
-            });
-
-            // With hex-line: 1 read_file call with paths:[] — concatenated output
-            const { value: withSL, ms: withMs } = runN(() => {
-                const parts = [];
-                for (const f of batchFiles) {
-                    parts.push(readFile(f));
-                }
-                return parts.join("\n\n---\n\n").length;
-            });
-
-            results.push({
-                num: 9, scenario: `Multi-file read (${batchFiles.length} files)`,
-                without, withSL,
-                savings: pctSavings(without, withSL),
-                latencyWithout: withoutMs, latencyWith: withMs,
-            });
-        }
-    }
-
-    // ===================================================================
-    // TEST 10: bulk_replace dry_run
-    // ===================================================================
-    {
-        const bulkTmpPaths = [];
-        for (let i = 0; i < 5; i++) {
-            const p = resolve(tmpdir(), `hex-line-bulk-${ts}-${i}.js`);
-            writeFileSync(p, tmpContent, "utf-8");
-            bulkTmpPaths.push(p);
-        }
-
-        const editLine = 13;
-        const editNew = '        this.configPath = resolve(configPath || ".");';
-
-        // Without hex-line: 5 separate edit_file calls
-        const { value: without, ms: withoutMs } = runN(() => {
-            let total = 0;
-            for (const p of bulkTmpPaths) {
-                const origLines = [...tmpLines];
-                const newLines = [...tmpLines];
-                newLines[editLine - 1] = editNew;
-                total += simBuiltInEdit(p, origLines, newLines).length;
-            }
-            return total;
-        });
-
-        // With hex-line: 1 bulk_replace — summary + per-file compact diff
-        const { value: withSL, ms: withMs } = runN(() => {
-            let response = "5 files changed, 0 errors\n";
-            for (const p of bulkTmpPaths) {
-                const origLines = [...tmpLines];
-                const newLines = [...tmpLines];
-                newLines[editLine - 1] = editNew;
-                response += simHexLineEditDiff(origLines, newLines) + "\n";
-            }
-            return response.length;
-        });
-
-        results.push({
-            num: 10, scenario: "bulk_replace dry_run (5 files)",
-            without, withSL,
-            savings: pctSavings(without, withSL),
-            latencyWithout: withoutMs, latencyWith: withMs,
-        });
-
-        for (const p of bulkTmpPaths) {
-            try { unlinkSync(p); } catch { /* ok */ }
-        }
-    }
-
-    // ===================================================================
-    // TEST 11: changes (semantic diff)
-    // ===================================================================
-    {
-        // Without hex-line: raw unified diff output
-        const { value: without, ms: withoutMs } = runN(() => {
-            const diffLines = [
-                `diff --git a/benchmark-target.js b/benchmark-target.js`,
-                `index abc1234..def5678 100644`,
-                `--- a/benchmark-target.js`,
-                `+++ b/benchmark-target.js`,
-                `@@ -10,6 +10,12 @@ const DEFAULT_TIMEOUT = 5000;`,
-            ];
-            // Simulate ~15 context + change lines typical of a small diff
-            for (let i = 0; i < 5; i++) {
-                diffLines.push(` ${tmpLines[i + 5] || "    // context line"}`);  // context
-            }
-            diffLines.push(`-${tmpLines[12] || "    old line"}`);
-            diffLines.push(`+        this.configPath = resolve(configPath || ".");`);
-            for (let i = 0; i < 5; i++) {
-                diffLines.push(` ${tmpLines[i + 14] || "    // context line"}`);  // context
-            }
-            // Second hunk — added function
-            diffLines.push(`@@ -195,0 +201,8 @@`);
-            for (let i = 0; i < 3; i++) {
-                diffLines.push(` ${tmpLines[i + 150] || "    // context"}`);
-            }
-            for (let i = 0; i < 5; i++) {
-                diffLines.push(`+    // new function line ${i}`);
-            }
-            for (let i = 0; i < 3; i++) {
-                diffLines.push(` ${tmpLines[i + 155] || "    // context"}`);
-            }
-            return diffLines.join("\n").length;
-        });
-
-        // With hex-line: real fileChanges() semantic diff (async, called once — deterministic)
-        let withSL;
-        let withMs = 0;
-        try {
-            const t0 = performance.now();
-            const changesOut = await fileChanges(allFiles[0]);
-            withMs = parseFloat((performance.now() - t0).toFixed(1));
-            withSL = changesOut.length;
-        } catch {
-            withSL = 133; // fallback if no git history
-        }
-
-        results.push({
-            num: 11, scenario: "Changes (semantic diff)",
-            without, withSL,
-            savings: pctSavings(without, withSL),
-            latencyWithout: withoutMs, latencyWith: withMs,
-        });
-    }
-
-    // ===================================================================
-    // TEST 12: FILE_NOT_FOUND recovery
-    // ===================================================================
-    {
-        const missingPath = resolve(repoRoot, "src/utils/halper.js");
-        const parentDir = resolve(repoRoot, "src/utils");
-
-        // Without hex-line: 3 round-trips (error → ls → retry)
-        const { value: without, ms: withoutMs } = runN(() => {
-            // Round 1: real ENOENT error
-            let r1;
-            try { readFileSync(missingPath, "utf-8"); r1 = ""; } catch (e) { r1 = e.message; }
-            // Round 2: real directory listing to find correct name
-            let r2;
-            try { r2 = readdirSync(parentDir).join("\n"); } catch { r2 = `${parentDir}: directory not found`; }
-            // Round 3: agent re-reads correct file (small file ~30 lines)
-            const r3 = simBuiltInReadFull(missingPath, tmpLines.slice(0, 30));
-            return (r1 + r2 + r3).length;
-        });
-
-        // With hex-line: real readFile() on nonexistent path — returns error + parent dir listing
-        const { value: withSL, ms: withMs } = runN(() => {
-            try {
-                return readFile(missingPath).length;
-            } catch (e) {
-                return e.message.length;
-            }
-        });
-
-        results.push({
-            num: 12, scenario: "FILE_NOT_FOUND recovery*",
-            without, withSL,
-            savings: pctSavings(without, withSL),
-            latencyWithout: withoutMs, latencyWith: withMs,
-        });
-    }
-
-    // ===================================================================
-    // TEST 13: Hash mismatch recovery
-    // ===================================================================
-    {
-        // Without hex-line: 3 round-trips (stale error → re-read full → retry edit)
-        const { value: without, ms: withoutMs } = runN(() => {
-            // Round 1: error
-            const r1 = 'Error: file content has changed (stale). Please re-read the file.';
-            // Round 2: full re-read
-            const r2 = simBuiltInReadFull(tmpPath, tmpLines);
-            // Round 3: retry edit response
-            const origLines = [...tmpLines];
-            const newLines = [...tmpLines];
-            newLines[12] = '        this.configPath = resolve(configPath || ".");';
-            const r3 = simBuiltInEdit(tmpPath, origLines, newLines);
-            return (r1 + r2 + r3).length;
-        });
-
-        // With hex-line: 1 round-trip (error + fresh snippet +/-5 lines around target)
-        const { value: withSL, ms: withMs } = runN(() => {
-            const targetLine = 13;
-            const snippetStart = Math.max(0, targetLine - 6);
-            const snippetEnd = Math.min(tmpLines.length, targetLine + 5);
-            const snippet = tmpLines.slice(snippetStart, snippetEnd);
-            const annotated = snippet.map((l, i) => {
-                const lineNum = snippetStart + i + 1;
-                const tag = lineTag(fnv1a(l));
-                return `${tag}.${lineNum}\t${l}`;
-            }).join("\n");
-            const response = `HASH_MISMATCH at line ${targetLine}. Fresh snippet:\n\`\`\`\n${annotated}\n\`\`\``;
-            return response.length;
-        });
-
-        results.push({
-            num: 13, scenario: "Hash mismatch recovery*",
-            without, withSL,
-            savings: pctSavings(without, withSL),
-            latencyWithout: withoutMs, latencyWith: withMs,
-        });
-    }
-
-    // ===================================================================
-    // TEST 14: Bash redirect savings
-    // ===================================================================
-    {
-        const infoFile = allFiles[Math.floor(allFiles.length / 2)] || allFiles[0];
-        const infoLines = getFileLines(infoFile);
-        if (infoLines) {
-            // Sub-test A: cat vs read_file
-            const catW = runN(() => {
-                // cat output: raw lines, no line numbers (agent redirect)
-                return infoLines.join("\n").length;
-            });
-            const catH = runN(() => readFile(infoFile).length);
-
-            // Sub-test B: ls -la vs directory_tree
-            const dirTarget = resolve(repoRoot);
-            const lsW = runN(() => simBuiltInLsR(dirTarget, 0, 1).length);
-            const lsH = runN(() => directoryTree(dirTarget, { max_depth: 1 }).length);
-
-            // Sub-test C: stat vs get_file_info
-            const stW = runN(() => simBuiltInStat(infoFile).length);
-            const stH = runN(() => fileInfo(infoFile).length);
-
-            // Combined: without = raw outputs (no follow-up possible)
-            // With = structured output (enables follow-up without extra calls)
-            const totalWithout = catW.value + lsW.value + stW.value;
-            const totalWith = catH.value + lsH.value + stH.value;
-            const totalMsWithout = catW.ms + lsW.ms + stW.ms;
-            const totalMsWith = catH.ms + lsH.ms + stH.ms;
-
-            results.push({
-                num: 14, scenario: "Bash redirects (cat+ls+stat)",
-                without: totalWithout, withSL: totalWith,
-                savings: pctSavings(totalWithout, totalWith),
-                latencyWithout: parseFloat(totalMsWithout.toFixed(1)), latencyWith: parseFloat(totalMsWith.toFixed(1)),
-            });
-        }
-    }
-
-    // ===================================================================
-    // TEST 15: HASH_HINT multi-match recovery
-    // ===================================================================
-    {
-        // Create a file with a duplicated line so textReplace triggers HASH_HINT
-        const dupLine = '    return this.config;';
-        const dupContent = tmpLines.map((l, i) => (i === 20 || i === 80) ? dupLine : l);
-        const dupPath = resolve(tmpdir(), `hex-line-dup-${ts}.js`);
-        writeFileSync(dupPath, dupContent.join("\n"), "utf-8");
-
-        // Without hex-line: 3 round-trips (opaque error + re-read full + retry)
-        const { value: without, ms: withoutMs } = runN(() => {
-            const r1 = 'Error: multiple occurrences found. Provide more context.';
-            const r2 = simBuiltInReadFull(dupPath, dupContent);
-            const origLines = [...dupContent];
-            const newLines = [...dupContent];
-            newLines[20] = '    return this.updatedConfig;';
-            const r3 = simBuiltInEdit(dupPath, origLines, newLines);
-            return (r1 + r2 + r3).length;
-        });
-
-        // With hex-line: HASH_HINT error contains annotated snippets (1 round-trip)
-        const { value: withSL, ms: withMs } = runN(() => {
-            try {
-                editFile(dupPath, [{ replace: { old_text: dupLine, new_text: '    return this.updatedConfig;' } }]);
-                return 0; // should not reach
-            } catch (e) {
-                // HASH_HINT error message + simulated anchor retry
-                const retry = '{"set_line":{"anchor":"xx.21","new_text":"    return this.updatedConfig;"}}';
-                return (e.message + retry).length;
-            }
-        });
-
-        results.push({
-            num: 15, scenario: "HASH_HINT multi-match recovery*",
-            without, withSL,
-            savings: pctSavings(without, withSL),
-            latencyWithout: withoutMs, latencyWith: withMs,
-        });
-
-        try { unlinkSync(dupPath); } catch { /* ok */ }
-    }
-
-    // ===================================================================
-    // TEST 16-18: Graph enrichment (--with-graph only)
-    // Both sides use hex-line; difference is whether .codegraph/index.db exists
-    // ===================================================================
-    const graphOut = [];
-    if (withGraph) {
-        const { getGraphDB, getRelativePath } = await import("./lib/graph-enrich.mjs");
-        const db = getGraphDB(resolve(repoRoot, "server.mjs"));
-        if (!db) {
-            console.error("--with-graph: .codegraph/index.db not found. Run hex-graph index_project first.");
-        } else {
-            const graphFile = largeFiles[0] || allFiles[0];
-            const graphLines = getFileLines(graphFile);
-
-            if (graphLines) {
-                // TEST 16: Read with/without Graph header
-                {
-                    const withGraphResult = readFile(graphFile);
-                    const noGraphResult = withGraphResult.replace(/\nGraph:.*\n/, "\n");
-                    const savings = pctSavings(noGraphResult.length, withGraphResult.length);
-                    graphOut.push(`| 16 | Graph: Read (${graphLines.length}L) | ${fmt(noGraphResult.length)} chars | ${fmt(withGraphResult.length)} chars | ${savings} | 2\u21921 | 2\u21921 |`);
-                }
-
-                // TEST 17: Edit with/without blast radius
-                {
-                    const editTmpPath = resolve(tmpdir(), `hex-bench-edit-${Date.now()}.js`);
-                    writeFileSync(editTmpPath, graphLines.join("\n"), "utf-8");
-                    try {
-                        const editResult = editFile(editTmpPath, [{ replace: { old_text: graphLines[5], new_text: graphLines[5] + " // modified" } }]);
-                        const noBlastOut = editResult.replace(/\n.*Blast radius.*$/s, "");
-                        const savings = pctSavings(noBlastOut.length, editResult.length);
-                        graphOut.push(`| 17 | Graph: Edit + impact | ${fmt(noBlastOut.length)} chars | ${fmt(editResult.length)} chars | ${savings} | 2\u21921 | 2\u21921 |`);
-                    } catch (e) {
-                        graphOut.push(`| 17 | Graph: Edit + impact | \u2014 | \u2014 | \u2014 | | |`);
-                    }
-                    try { unlinkSync(editTmpPath); } catch {}
-                }
-
-                // TEST 18: Grep with/without annotations
-                {
-                    try {
-                        const grepResult = await grepSearch("function", { path: resolve(repoRoot), glob: "*.mjs", limit: 10 });
-                        const noAnnoResult = grepResult.replace(/  \[[^\]]+\]/g, "");
-                        const savings = pctSavings(noAnnoResult.length, grepResult.length);
-                        const annoCount = (grepResult.match(/\[[^\]]+\]/g) || []).length;
-                        graphOut.push(`| 18 | Graph: Grep + ${annoCount} annotations | ${fmt(noAnnoResult.length)} chars | ${fmt(grepResult.length)} chars | ${savings} | 6\u21921 | 6\u21921 |`);
-                    } catch {
-                        graphOut.push(`| 18 | Graph: Grep + context | \u2014 | \u2014 | \u2014 | | |`);
-                    }
-                }
-            }
-        }
-    }
-
-    // ===================================================================
-    // WORKFLOW SCENARIOS (multi-step real operations)
-    // ===================================================================
-    const workflowResults = [];
-
-    // W1: Search → Edit (find a pattern, edit the match)
-    {
-        const wTmpPath = resolve(tmpdir(), `hex-wf1-${Date.now()}.js`);
-        writeFileSync(wTmpPath, tmpContent, "utf-8");
-        const editLine = tmpLines[12];
-        const editNew = editLine + " // workflow-modified";
-
-        // Without: grep → read file for context → edit with old_string
-        const { value: without } = runN(() => {
-            let total = 0;
-            // Step 1: grep to find
-            total += simBuiltInGrep("configPath", wTmpPath).length;
-            // Step 2: read full file for context (agent needs surrounding lines)
-            total += simBuiltInReadFull(wTmpPath, tmpLines).length;
-            // Step 3: edit
-            const origLines = [...tmpLines];
-            const newLines = [...tmpLines];
-            newLines[12] = editNew;
-            total += simBuiltInEdit(wTmpPath, origLines, newLines).length;
-            return total;
-        });
-
-        // With: grep_search (has hashes) → edit with anchor (no re-read needed)
-        const { value: withSL } = runN(() => {
-            let total = 0;
-            // Step 1: grep with hashes
-            const grepOut = readFileSync(wTmpPath, "utf-8"); // simulate grep result
-            const lines = grepOut.split("\n");
-            const targetIdx = 12;
-            const tag = lineTag(fnv1a(lines[targetIdx]));
-            total += `${wTmpPath}:>>${tag}.${targetIdx + 1}\t${lines[targetIdx]}`.length;
-            // Step 2: edit with anchor directly (no read needed)
-            try {
-                const result = editFile(wTmpPath, [{ set_line: { anchor: `${tag}.${targetIdx + 1}`, new_text: editNew } }]);
-                total += result.length;
-            } catch (e) { total += e.message.length; }
-            return total;
-        });
-
-        workflowResults.push({
-            id: "W1", scenario: "Search \u2192 Edit",
-            without, withSL,
-            opsWithout: 3, opsWith: 2,
-        });
-        try { unlinkSync(wTmpPath); } catch {}
-    }
-
-    // W2: Read → Edit → Verify cycle
-    {
-        const wTmpPath = resolve(tmpdir(), `hex-wf2-${Date.now()}.js`);
-        writeFileSync(wTmpPath, tmpContent, "utf-8");
-
-        // Without: read full → edit → re-read full to verify
-        const { value: without } = runN(() => {
-            let total = 0;
-            total += simBuiltInReadFull(wTmpPath, tmpLines).length; // read
-            const origLines = [...tmpLines];
-            const newLines = [...tmpLines];
-            newLines[12] = '        this.configPath = resolve(configPath || ".");';
-            total += simBuiltInEdit(wTmpPath, origLines, newLines).length; // edit
-            total += simBuiltInReadFull(wTmpPath, tmpLines).length; // re-read to verify
-            return total;
-        });
-
-        // With: read targeted → edit → verify checksums
-        const { value: withSL } = runN(() => {
-            let total = 0;
-            total += readFile(wTmpPath, { offset: 8, limit: 20 }).length; // targeted read
-            // Reset file for edit
-            writeFileSync(wTmpPath, tmpContent, "utf-8");
-            try {
-                const result = editFile(wTmpPath, [{ replace: { old_text: tmpLines[12], new_text: '        this.configPath = resolve(configPath || ".");' } }]);
-                total += result.length;
-            } catch (e) { total += e.message.length; }
-            // Verify with checksums instead of re-reading
-            const hashes = tmpLines.slice(0, 50).map(l => fnv1a(l));
-            const cs = rangeChecksum(hashes, 1, 50);
-            try { total += verifyChecksums(wTmpPath, [cs]).length; }
-            catch { total += 100; }
-            return total;
-        });
-
-        workflowResults.push({
-            id: "W2", scenario: "Read \u2192 Edit \u2192 Verify",
-            without, withSL,
-            opsWithout: 3, opsWith: 3,
-        });
-        try { unlinkSync(wTmpPath); } catch {}
-    }
-
-    // W3: Multi-file refactor (rename in 5 files)
-    {
-        const wDir = resolve(tmpdir(), `hex-wf3-${Date.now()}`);
-        mkdirSync(wDir, { recursive: true });
-        const wPaths = [];
-        for (let i = 0; i < 5; i++) {
-            const p = resolve(wDir, `file-${i}.js`);
-            writeFileSync(p, tmpContent, "utf-8");
-            wPaths.push(p);
-        }
-
-        // Without: grep to find files → read each → edit each = 11 ops
-        const { value: without } = runN(() => {
-            let total = 0;
-            total += simBuiltInGrep("configPath", wPaths[0]).length; // find
-            for (const p of wPaths) {
-                total += simBuiltInReadFull(p, tmpLines).length; // read each
-                const origLines = [...tmpLines];
-                const newLines = [...tmpLines];
-                newLines[12] = newLines[12].replace("configPath", "settingsPath");
-                total += simBuiltInEdit(p, origLines, newLines).length; // edit each
-            }
-            return total;
-        });
-
-        // With: grep_search → bulk_replace = 2 ops
-        const { value: withSL } = runN(() => {
-            let total = 0;
-            // Restore files
-            for (const p of wPaths) writeFileSync(p, tmpContent, "utf-8");
-            // Single grep (simulated — bulk_replace does its own finding)
-            total += 200; // approximate grep output
-            // Single bulk_replace
-            const result = bulkReplace(
-                wDir,
-                "*.js",
-                [{ old: "configPath", new: "settingsPath" }],
-                { dryRun: true, maxFiles: 10 }
-            );
-            total += result.length;
-            return total;
-        });
-
-        workflowResults.push({
-            id: "W3", scenario: "Multi-file refactor (5 files)",
-            without, withSL,
-            opsWithout: 11, opsWith: 2,
-        });
-        try { rmSync(wDir, { recursive: true }); } catch {}
-    }
-
-    // W4: Explore large file → targeted edit
-    {
-        const largeFile = largeFiles[0] || allFiles[0];
-        const largeLines = getFileLines(largeFile);
-        if (largeLines && largeLines.length > 100) {
-            // Without: read full file → grep for method → edit
-            const { value: without } = runN(() => {
-                let total = 0;
-                total += simBuiltInReadFull(largeFile, largeLines).length;
-                total += simBuiltInGrep("function", largeFile).length;
-                // Simulate edit response
-                const origLines = [...largeLines];
-                const newLines = [...largeLines];
-                newLines[10] = newLines[10] + " // modified";
-                total += simBuiltInEdit(largeFile, origLines, newLines).length;
-                return total;
-            });
-
-            // With: outline → read range → edit with anchor
-            let outlineLen = 500;
-            try { outlineLen = (await fileOutline(largeFile)).length; } catch {}
-            const { value: withSL } = runN(() => {
-                let total = 0;
-                total += outlineLen; // outline (pre-computed, async)
-                total += readFile(largeFile, { offset: 5, limit: 30 }).length; // targeted read
-                total += simHexLineEditDiff(largeLines.slice(5, 35), [...largeLines.slice(5, 35)].map((l, i) => i === 5 ? l + " // modified" : l)).length;
-                return total;
-            });
-
-            workflowResults.push({
-                id: "W4", scenario: `Explore+edit (${largeLines.length}L file)`,
-                without, withSL,
-                opsWithout: 3, opsWith: 3,
-            });
-        }
-    }
-
-    // ===================================================================
-    // Cleanup
-    // ===================================================================
-    try { unlinkSync(tmpPath); } catch { /* ok */ }
-
-    // ===================================================================
-    // Report
-    // ===================================================================
-    const out = [];
-    out.push("# Hex-line Benchmark v3");
-    out.push("");
-    out.push(`Repository: ${repoName} (${fmt(allFiles.length)} code files, ${fmt(totalLines)} lines)  `);
-    out.push(`Temp file: ${tmpPath} (200 lines)  `);
-    out.push(`Date: ${new Date().toISOString().slice(0, 10)}  `);
-    out.push(`Runs per scenario: ${RUNS} (median)  `);
-    out.push("");
-
-    // Ops comparison: how many tool calls each scenario requires
-    const OPS = {
-        "Read full (<50L)":           { without: 1, with: 1 },
-        "Read full (50-200L)":        { without: 1, with: 1 },
-        "Read full (200-500L)":       { without: 1, with: 1 },
-        "Read full (500L+)":          { without: 1, with: 1 },
-        "Outline+read (200-500L)":    { without: 1, with: 2 },
-        "Outline+read (500L+)":       { without: 1, with: 2 },
-        "Grep search":                { without: 1, with: 1 },
-        "Directory tree":             { without: 1, with: 1 },
-        "File info":                  { without: 1, with: 1 },
-        "Create file (200L)":         { without: 1, with: 1 },
-        "Edit x5 sequential":         { without: 5, with: 5 },
-        "Verify checksums (4 ranges)": { without: 4, with: 1 },
-        "Multi-file read":            { without: 2, with: 1 },
-        "bulk_replace dry_run (5 files)": { without: 5, with: 1 },
-        "Changes (semantic diff)":    { without: 1, with: 1 },
-        "FILE_NOT_FOUND recovery*":   { without: 3, with: 1 },
-        "Hash mismatch recovery*":    { without: 3, with: 1 },
-        "Bash redirects (cat+ls+stat)": { without: 3, with: 3 },
-        "HASH_HINT multi-match recovery*": { without: 3, with: 2 },
-    };
-
-    const STEPS = {
-        "Read full (<50L)": { without: 1, with: 1 },
-        "Read full (50-200L)": { without: 1, with: 1 },
-        "Read full (200-500L)": { without: 1, with: 1 },
-        "Read full (500L+)": { without: 1, with: 1 },
-        "Outline+read (200-500L)": { without: 1, with: 2 },
-        "Outline+read (500L+)": { without: 1, with: 2 },
-        "Grep search": { without: 1, with: 1 },
-        "Directory tree": { without: 1, with: 1 },
-        "File info": { without: 1, with: 1 },
-        "Create file (200L)": { without: 1, with: 1 },
-        "Edit x5 sequential": { without: 5, with: 5 },
-        "Verify checksums (4 ranges)": { without: 4, with: 1 },
-        "Multi-file read": { without: 1, with: 1 },
-        "bulk_replace dry_run (5 files)": { without: 5, with: 1 },
-        "Changes (semantic diff)": { without: 1, with: 1 },
-        "FILE_NOT_FOUND recovery": { without: 3, with: 1 },
-        "Hash mismatch recovery": { without: 3, with: 1 },
-        "Bash redirects (cat+ls+stat)": { without: 1, with: 1 },
-        "HASH_HINT multi-match recovery": { without: 3, with: 1 },
-    };
-
-    // Combined results + ops + steps table
-    out.push("## Results");
-    out.push("");
-    out.push("| # | Scenario | Baseline | Hex-line | Savings | Ops | Steps |");
-    out.push("|---|----------|----------|----------|---------|-----|-------|");
-
-    for (const r of results) {
-        if (r.num >= 16) continue; // graph rows added below
-
-        // Match OPS/STEPS keys
-        let op = OPS[r.scenario];
-        if (!op) {
-            const key = Object.keys(OPS).find(k => r.scenario.startsWith(k));
-            if (key) op = OPS[key];
-        }
-        let step = STEPS[r.scenario];
-        if (!step) {
-            const key = Object.keys(STEPS).find(k => r.scenario.startsWith(k));
-            if (key) step = STEPS[key];
-        }
-
-        const opsStr = op ? `${op.without}\u2192${op.with}` : "\u2014";
-        const stepsStr = step ? `${step.without}\u2192${step.with}` : "\u2014";
-
-        out.push(`| ${r.num} | ${r.scenario} | ${fmt(r.without)} chars | ${fmt(r.withSL)} chars | ${r.savings} | ${opsStr} | ${stepsStr} |`);
-    }
-
-    // Append graph rows into same table (if any)
-    if (graphOut.length > 0) {
-        out.push("| | **hex-line \u00b1 graph** | **No Graph** | **With Graph** | | | |");
-        out.push(...graphOut);
-    }
-    out.push("");
-
-    // Workflow scenarios table
-    if (workflowResults.length > 0) {
-        out.push("## Workflow Scenarios (multi-step)");
-        out.push("");
-        out.push("| # | Scenario | Built-in | Hex-line | Savings | Ops |");
-        out.push("|---|----------|----------|----------|---------|-----|");
-        for (const w of workflowResults) {
-            out.push(`| ${w.id} | ${w.scenario} | ${fmt(w.without)} chars | ${fmt(w.withSL)} chars | ${pctSavings(w.without, w.withSL)} | ${w.opsWithout}\u2192${w.opsWith} |`);
-        }
-        out.push("");
-    }
-
-    // Verdict
-    out.push("## Verdict");
-    out.push("");
-
-    const readResults = results.filter(r => r.num === 1);
-    const outlineResults = results.filter(r => r.num === 2);
-    const editResult = results.find(r => r.num === 7);
-    const verifyResult = results.find(r => r.num === 8);
-    const treeResult = results.find(r => r.num === 4);
-    const batchResult = results.find(r => r.num === 9);
-    const bulkResult = results.find(r => r.num === 10);
-    const changesResult = results.find(r => r.num === 11);
-    const notFoundResult = results.find(r => r.num === 12);
-    const mismatchResult = results.find(r => r.num === 13);
-    const bashResult = results.find(r => r.num === 14);
-
-    const coreResults = results.filter(r => r.num < 16);
-    const allSavingsNums = coreResults.map(r => {
-        if (r.without === 0) return 0;
-        return ((r.without - r.withSL) / r.without) * 100;
-    });
-    const avgSavings = allSavingsNums.length > 0
-        ? allSavingsNums.reduce((a, b) => a + b, 0) / allSavingsNums.length
-        : 0;
-
-    // Weighted average based on typical development session frequency
-    const WEIGHTS = {
-        "Read full (<50L)": 2, "Read full (50-200L)": 5, "Read full (200-500L)": 3, "Read full (500L+)": 1,
-        "Outline+read (200-500L)": 8, "Outline+read (500L+)": 8,
-        "Grep search": 5, "Directory tree": 2, "File info": 1, "Create file (200L)": 1,
-        "Edit x5 sequential": 10, "Verify checksums (4 ranges)": 8,
-        "Multi-file read": 2, "bulk_replace dry_run (5 files)": 1,
-        "Changes (semantic diff)": 3,
-        "FILE_NOT_FOUND recovery": 2, "Hash mismatch recovery": 3,
-        "Bash redirects (cat+ls+stat)": 3, "HASH_HINT multi-match recovery": 2,
-    };
-    let wSum = 0, wTotal = 0;
-    for (const r of coreResults) {
-        const w = WEIGHTS[r.scenario] || 1;
-        const sav = r.without === 0 ? 0 : ((r.without - r.withSL) / r.without) * 100;
-        wSum += w * sav;
-        wTotal += w;
-    }
-    const weightedAvg = wTotal > 0 ? wSum / wTotal : 0;
-
-    // Ops/Steps totals for core scenarios
-    const totalOpsWithout = coreResults.reduce((s, r) => {
-        let op = OPS[r.scenario];
-        if (!op) { const key = Object.keys(OPS).find(k => r.scenario.startsWith(k)); if (key) op = OPS[key]; }
-        return s + (op ? op.without : 1);
-    }, 0);
-    const totalOpsWith = coreResults.reduce((s, r) => {
-        let op = OPS[r.scenario];
-        if (!op) { const key = Object.keys(OPS).find(k => r.scenario.startsWith(k)); if (key) op = OPS[key]; }
-        return s + (op ? op.with : 1);
-    }, 0);
-    const totalStepsWithout = coreResults.reduce((s, r) => {
-        let step = STEPS[r.scenario];
-        if (!step) { const key = Object.keys(STEPS).find(k => r.scenario.startsWith(k)); if (key) step = STEPS[key]; }
-        return s + (step ? step.without : 1);
-    }, 0);
-    const totalStepsWith = coreResults.reduce((s, r) => {
-        let step = STEPS[r.scenario];
-        if (!step) { const key = Object.keys(STEPS).find(k => r.scenario.startsWith(k)); if (key) step = STEPS[key]; }
-        return s + (step ? step.with : 1);
-    }, 0);
-    const opsPct = totalOpsWithout > 0 ? ((totalOpsWithout - totalOpsWith) / totalOpsWithout * 100).toFixed(0) : 0;
-
-    // Read verdict
-    const readVerdict = [];
-    const smallRead = readResults.find(r => r.scenario.includes("<50L"));
-    const xlRead = readResults.find(r => r.scenario.includes("500L+"));
-    if (smallRead) {
-        const pct = Math.abs(((smallRead.without - smallRead.withSL) / smallRead.without * 100)).toFixed(0);
-        const verb = smallRead.withSL <= smallRead.without ? "saves" : "costs";
-        readVerdict.push(`Small files (<50L): hash annotations ${verb} ~${pct}%.`);
-    }
-    if (xlRead) {
-        const pct = Math.abs(((xlRead.without - xlRead.withSL) / xlRead.without * 100)).toFixed(0);
-        const verb = xlRead.withSL <= xlRead.without ? "saves" : "costs";
-        readVerdict.push(`Large files (500L+): full read ${verb} ~${pct}%.`);
-    }
-
-    out.push("**Read:**");
-    for (const v of readVerdict) out.push(`- ${v}`);
-    if (outlineResults.length > 0) {
-        const best = outlineResults.reduce((a, b) =>
-            ((a.without - a.withSL) / a.without) > ((b.without - b.withSL) / b.without) ? a : b
-        );
-        const savPct = ((best.without - best.withSL) / best.without * 100).toFixed(0);
-        out.push(`- Outline+targeted read saves ${savPct}% on large files vs full read.`);
-    }
-    out.push("");
-
-    if (editResult) {
-        const editSav = ((editResult.without - editResult.withSL) / editResult.without * 100).toFixed(0);
-        out.push(`**Edit:** Compact diff output saves ${editSav}% vs old_string/new_string context blocks (5 edits).`);
-        out.push("");
-    }
-
-    if (verifyResult) {
-        const verifySav = ((verifyResult.without - verifyResult.withSL) / verifyResult.without * 100).toFixed(0);
-        out.push(`**Verify:** Checksum verification saves ${verifySav}% vs full re-read for staleness check.`);
-        out.push("");
-    }
-
-    if (treeResult) {
-        const pct = Math.abs(((treeResult.without - treeResult.withSL) / treeResult.without * 100)).toFixed(0);
-        const verb = treeResult.withSL <= treeResult.without ? "saves" : "costs";
-        out.push(`**Tree:** Compact directory tree ${verb} ${pct}% vs \`ls -laR\`.`);
-        out.push("");
-    }
-
-    if (batchResult) {
-        const batchSav = ((batchResult.without - batchResult.withSL) / batchResult.without * 100).toFixed(0);
-        out.push(`**Batch read:** Multi-file read saves ${batchSav}% vs separate Read calls (${batchResult.scenario.match(/\d+ files/)?.[0] || 'N files'}).`);
-        out.push("");
-    }
-
-    if (bulkResult) {
-        const bulkSav = ((bulkResult.without - bulkResult.withSL) / bulkResult.without * 100).toFixed(0);
-        out.push(`**Bulk replace:** Single bulk_replace saves ${bulkSav}% vs 5 separate edit_file calls.`);
-        out.push("");
-    }
-
-    if (changesResult) {
-        const changesSav = ((changesResult.without - changesResult.withSL) / changesResult.without * 100).toFixed(0);
-        out.push(`**Changes:** Semantic diff summary saves ${changesSav}% vs raw unified diff output.`);
-        out.push("");
-    }
-
-    if (notFoundResult) {
-        const notFoundSav = ((notFoundResult.without - notFoundResult.withSL) / notFoundResult.without * 100).toFixed(0);
-        out.push(`**Error recovery (FILE_NOT_FOUND):** Inline dir listing saves ${notFoundSav}% vs 3 round-trips.`);
-        out.push("");
-    }
-
-    if (mismatchResult) {
-        const mismatchSav = ((mismatchResult.without - mismatchResult.withSL) / mismatchResult.without * 100).toFixed(0);
-        out.push(`**Error recovery (hash mismatch):** Fresh snippet saves ${mismatchSav}% vs full re-read + retry.`);
-        out.push("");
-    }
-
-    if (bashResult) {
-        const bashSav = ((bashResult.without - bashResult.withSL) / bashResult.without * 100).toFixed(0);
-        const verb = bashResult.withSL <= bashResult.without ? "saves" : "costs";
-        out.push(`**Bash redirects:** Structured hex-line output ${verb} ${bashSav}% vs cat+ls+stat combined.`);
-        out.push("");
-    }
-
-    // Break-even
-    out.push("## Break-even");
-    out.push("");
-    if (outlineResults.length > 0) {
-        out.push("- **Outline workflow** breaks even at ~50 lines. Above that, savings grow linearly.");
-    }
-    if (verifyResult && verifyResult.withSL < verifyResult.without) {
-        const ratio = (verifyResult.without / verifyResult.withSL).toFixed(0);
-        out.push(`- **Verify** is ${ratio}x cheaper than re-reading. Pays for hash overhead after first staleness check.`);
-    }
-    if (editResult && editResult.withSL < editResult.without) {
-        out.push("- **Edit** compact diff is always cheaper than old_string/new_string blocks.");
-    }
-    if (notFoundResult && notFoundResult.withSL < notFoundResult.without) {
-        const ratio = (notFoundResult.without / notFoundResult.withSL).toFixed(0);
-        out.push(`- **Error recovery** eliminates round-trips: ${ratio}x cheaper for FILE_NOT_FOUND.`);
-    }
-    if (mismatchResult && mismatchResult.withSL < mismatchResult.without) {
-        const ratio = (mismatchResult.without / mismatchResult.withSL).toFixed(0);
-        out.push(`- **Hash mismatch** recovery with fresh snippet is ${ratio}x cheaper than full re-read + retry.`);
-    }
-    if (changesResult && changesResult.withSL < changesResult.without) {
-        out.push("- **Semantic diff** always cheaper than raw unified diff for understanding changes.");
-    }
-    out.push(`- **Average:** ${avgSavings.toFixed(0)}% tokens (flat) / ${weightedAvg.toFixed(0)}% (weighted) | ${totalOpsWithout}\u2192${totalOpsWith} ops (${opsPct}% fewer) | ${totalStepsWithout}\u2192${totalStepsWith} steps`);
-    out.push("");
-
-    console.log(out.join("\n"));
-}
-
-main();