diffdoc 0.6.2 → 0.6.4

package/.diffdocrc.example

@@ -1,17 +1,19 @@
-{
-  "baseDir": "./.diffdoc",
-  "aiProvider": "local",
-  "localLlmEndpoint": "http://localhost:11434/v1",
-  "localEmbedEndpoint": "http://localhost:11434/v1/embeddings",
-  "localChatModel": "qwen2.5-coder:7b",
-  "localEmbedModel": "nomic-embed-code",
-  "cloudLlmEndpoint": "https://api.openai.com/v1",
-  "cloudChatModel": "gpt-4o-mini",
-  "cloudEmbedModel": "text-embedding-3-small",
-  "embedBatchSize": 25,
-  "summarizeConcurrency": 2,
-  "openaiApiKey": "",
-  "includeGlobs": [],
-  "excludeGlobs": [],
-  "ignoreFile": ".diffdocignore"
-}
+{
+  "$schema": "https://raw.githubusercontent.com/sullyTheDev/diffdoc/main/schemas/v2/diffdocrc.schema.json",
+  "baseDir": "./.diffdoc",
+  "repoPath": ".",
+  "aiProvider": "local",
+  "localLlmEndpoint": "http://localhost:11434/v1",
+  "localEmbedEndpoint": "http://localhost:11434/v1/embeddings",
+  "localChatModel": "qwen2.5-coder:7b",
+  "localEmbedModel": "nomic-embed-code",
+  "cloudLlmEndpoint": "https://api.openai.com/v1",
+  "cloudChatModel": "gpt-4o-mini",
+  "cloudEmbedModel": "text-embedding-3-small",
+  "embedBatchSize": 25,
+  "summarizeConcurrency": 2,
+  "openaiApiKey": "",
+  "includeGlobs": [],
+  "excludeGlobs": [],
+  "ignoreFile": ".diffdocignore"
+}

package/dist/commands/prune.js

@@ -0,0 +1,186 @@
+"use strict";
+var __importDefault = (this && this.__importDefault) || function (mod) {
+    return (mod && mod.__esModule) ? mod : { "default": mod };
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.runPrune = runPrune;
+const promises_1 = __importDefault(require("node:fs/promises"));
+const node_path_1 = __importDefault(require("node:path"));
+const promises_2 = require("node:readline/promises");
+const node_process_1 = require("node:process");
+const artifacts_1 = require("../types/artifacts");
+const paths_1 = require("../utils/paths");
+const scan_1 = require("../utils/scan");
+function getSummaryDir(manifestPath) {
+    return node_path_1.default.resolve(node_path_1.default.dirname(manifestPath), "summaries");
+}
+function getSummaryPath(summaryDir, hash) {
+    return node_path_1.default.resolve(summaryDir, `${hash}.json`);
+}
+async function readManifest(manifestPath) {
+    const raw = JSON.parse(await promises_1.default.readFile(manifestPath, "utf8"));
+    const result = artifacts_1.RepoManifestSchema.safeParse(raw);
+    if (!result.success) {
+        const issues = result.error.issues.map((i) => `  - ${i.path.join(".")}: ${i.message}`).join("\n");
+        throw new Error(`Invalid manifest in ${manifestPath}:\n${issues}`);
+    }
+    return result.data;
+}
+async function writeManifest(manifestPath, manifest) {
+    await promises_1.default.mkdir(node_path_1.default.dirname(manifestPath), { recursive: true });
+    await promises_1.default.writeFile(manifestPath, `${JSON.stringify(manifest, null, 2)}\n`);
+}
+function countHashRefs(files) {
+    const refs = new Map();
+    for (const hash of Object.values(files)) {
+        refs.set(hash, (refs.get(hash) || 0) + 1);
+    }
+    return refs;
+}
+async function fileExists(filePath) {
+    try {
+        await promises_1.default.access(filePath);
+        return true;
+    }
+    catch {
+        return false;
+    }
+}
+async function promptConfirm(message) {
+    const rl = (0, promises_2.createInterface)({ input: node_process_1.stdin, output: node_process_1.stdout });
+    try {
+        const answer = await rl.question(`${message} (y/N): `);
+        return answer.trim().toLowerCase() === "y";
+    }
+    finally {
+        rl.close();
+    }
+}
+async function runPrune(options, config) {
+    const commandCwd = process.cwd();
+    const repoPath = node_path_1.default.resolve(commandCwd, options.path || config.repoPath);
+    const manifestPath = (0, paths_1.resolveDiffdocArtifactPath)(options.manifest, config.baseDir);
+    const summaryDir = getSummaryDir(manifestPath);
+    const manifest = await readManifest(manifestPath);
+    const manifestFilePaths = Object.keys(manifest.files);
+    // Scan eligible files
+    const includePatterns = (0, scan_1.compileGlobs)(config.summarize.includeGlobs.map(scan_1.normalizeGlobPattern));
+    const excludePatterns = (0, scan_1.compileGlobs)(config.summarize.excludeGlobs.map(scan_1.normalizeGlobPattern));
+    const ignoreMatcher = await (0, scan_1.readIgnoreMatcher)(repoPath, config.summarize.ignoreFile);
+    const scannedFiles = await (0, scan_1.walkCodeFiles)(repoPath, includePatterns, excludePatterns, ignoreMatcher);
+    const scannedFileSet = new Set(scannedFiles);
+    // Identify entries to prune
+    const toPrune = [];
+    for (const filePath of manifestFilePaths) {
+        if (scannedFileSet.has(filePath)) {
+            continue;
+        }
+        // Check if the file is deleted from disk or just excluded by rules
+        const absolutePath = node_path_1.default.resolve(repoPath, filePath);
+        const exists = await fileExists(absolutePath);
+        toPrune.push({
+            filePath,
+            reason: exists ? "excluded" : "deleted"
+        });
+    }
+    if (toPrune.length === 0) {
+        if (options.json) {
+            const report = {
+                scannedFileCount: scannedFiles.length,
+                manifestFileCount: manifestFilePaths.length,
+                pruned: [],
+                manifestEntriesRemoved: 0,
+                summaryAssetsDeleted: 0
+            };
+            console.log(JSON.stringify(report, null, 2));
+        }
+        else {
+            console.log(`Scanned files: ${scannedFiles.length}`);
+            console.log(`Manifest files: ${manifestFilePaths.length}`);
+            console.log("Nothing to prune.");
+        }
+        return;
+    }
+    // Dry run: report and exit
+    if (options.dryRun) {
+        if (options.json) {
+            const report = {
+                scannedFileCount: scannedFiles.length,
+                manifestFileCount: manifestFilePaths.length,
+                pruned: toPrune,
+                manifestEntriesRemoved: toPrune.length,
+                summaryAssetsDeleted: 0 // unknown until actual execution
+            };
+            console.log(JSON.stringify(report, null, 2));
+        }
+        else {
+            console.log(`Scanned files: ${scannedFiles.length}`);
+            console.log(`Manifest files: ${manifestFilePaths.length}`);
+            console.log(`Files to prune: ${toPrune.length}`);
+            for (const entry of toPrune) {
+                console.log(`  - ${entry.filePath} (${entry.reason})`);
+            }
+            console.log("");
+            console.log("Dry run complete. Use without --dry-run to execute.");
+        }
+        return;
+    }
+    // Confirmation prompt
+    if (!options.yes) {
+        if (!options.json) {
+            console.log(`Scanned files: ${scannedFiles.length}`);
+            console.log(`Manifest files: ${manifestFilePaths.length}`);
+            console.log(`Files to prune: ${toPrune.length}`);
+            for (const entry of toPrune) {
+                console.log(`  - ${entry.filePath} (${entry.reason})`);
+            }
+            console.log("");
+        }
+        const confirmed = await promptConfirm(`Remove ${toPrune.length} entries from manifest?`);
+        if (!confirmed) {
+            console.log("Aborted.");
+            return;
+        }
+    }
+    // Execute pruning
+    const refs = countHashRefs(manifest.files);
+    let summaryAssetsDeleted = 0;
+    for (const entry of toPrune) {
+        const hash = manifest.files[entry.filePath];
+        if (!hash) {
+            continue;
+        }
+        delete manifest.files[entry.filePath];
+        const newRefCount = Math.max((refs.get(hash) || 1) - 1, 0);
+        refs.set(hash, newRefCount);
+        // Delete summary asset if no longer referenced
+        if (newRefCount === 0) {
+            const summaryPath = getSummaryPath(summaryDir, hash);
+            try {
+                await promises_1.default.unlink(summaryPath);
+                summaryAssetsDeleted += 1;
+            }
+            catch (error) {
+                const nodeError = error;
+                if (nodeError.code !== "ENOENT") {
+                    throw error;
+                }
+            }
+        }
+    }
+    await writeManifest(manifestPath, manifest);
+    const report = {
+        scannedFileCount: scannedFiles.length,
+        manifestFileCount: manifestFilePaths.length,
+        pruned: toPrune,
+        manifestEntriesRemoved: toPrune.length,
+        summaryAssetsDeleted
+    };
+    if (options.json) {
+        console.log(JSON.stringify(report, null, 2));
+    }
+    else {
+        console.log(`Removed ${toPrune.length} manifest entries.`);
+        console.log(`Deleted ${summaryAssetsDeleted} orphaned summary assets.`);
+    }
+}

package/dist/commands/status.js

@@ -11,6 +11,7 @@ const embed_1 = require("./embed");
 const artifacts_1 = require("../types/artifacts");
 const paths_1 = require("../utils/paths");
 const llm_1 = require("../utils/llm");
+const scan_1 = require("../utils/scan");
 function getSummaryDir(manifestPath) {
     return node_path_1.default.resolve(node_path_1.default.dirname(manifestPath), "summaries");
 }
@@ -140,6 +141,12 @@ async function getIndexFreshness(manifest, config) {
         extra
     };
 }
+async function scanEligibleFiles(repoPath, config) {
+    const includePatterns = (0, scan_1.compileGlobs)(config.summarize.includeGlobs.map(scan_1.normalizeGlobPattern));
+    const excludePatterns = (0, scan_1.compileGlobs)(config.summarize.excludeGlobs.map(scan_1.normalizeGlobPattern));
+    const ignoreMatcher = await (0, scan_1.readIgnoreMatcher)(repoPath, config.summarize.ignoreFile);
+    return (0, scan_1.walkCodeFiles)(repoPath, includePatterns, excludePatterns, ignoreMatcher);
+}
 function formatSummaryFreshness(stats) {
     if (stats.missingFromManifestCount === 0 && stats.staleCount === 0) {
         return "fresh";
@@ -147,6 +154,10 @@ function formatSummaryFreshness(stats) {
     return `stale (missing: ${stats.missingFromManifestCount}, stale: ${stats.staleCount})`;
 }
 function buildSummarizeCommand(manifestOption) {
+    const command = "diffdoc summarize --mode delta";
+    return manifestOption === "manifest.json" ? command : `${command} --out ${manifestOption}`;
+}
+function buildSummarizeRefreshCommand(manifestOption) {
     const command = "diffdoc summarize --mode all --refresh";
     return manifestOption === "manifest.json" ? command : `${command} --out ${manifestOption}`;
 }
@@ -154,10 +165,22 @@ function buildEmbedCommand(manifestOption) {
     const command = "diffdoc embed";
     return manifestOption === "manifest.json" ? command : `${command} --manifest ${manifestOption}`;
 }
-function getNextCommand(manifestOption, summaryStats, indexFreshness) {
-    if (summaryStats.missingFromManifestCount > 0 || summaryStats.staleCount > 0) {
+function getNextCommand(manifestOption, summaryStats, indexFreshness, untrackedCount, newlyExcludedCount) {
+    if (newlyExcludedCount > 0) {
+        return {
+            command: "diffdoc prune",
+            reason: `${newlyExcludedCount} file(s) in manifest are no longer eligible`
+        };
+    }
+    if (untrackedCount > 0) {
         return {
             command: buildSummarizeCommand(manifestOption),
+            reason: `${untrackedCount} eligible file(s) not yet summarized`
+        };
+    }
+    if (summaryStats.missingFromManifestCount > 0 || summaryStats.staleCount > 0) {
+        return {
+            command: buildSummarizeRefreshCommand(manifestOption),
             reason: "summary artifacts are missing or stale"
         };
     }
@@ -178,11 +201,18 @@ function getNextCommand(manifestOption, summaryStats, indexFreshness) {
         reason: "summaries and index are fresh"
     };
 }
-function buildStatusReport(manifest, summaryStats, indexFreshness, manifestOption) {
-    const nextCommand = getNextCommand(manifestOption, summaryStats, indexFreshness);
+function buildStatusReport(manifest, scannedFiles, summaryStats, indexFreshness, manifestOption) {
+    const manifestFilePaths = new Set(Object.keys(manifest.files));
+    const scannedFileSet = new Set(scannedFiles);
+    const untrackedCount = scannedFiles.filter((f) => !manifestFilePaths.has(f)).length;
+    const newlyExcludedCount = [...manifestFilePaths].filter((f) => !scannedFileSet.has(f)).length;
+    const nextCommand = getNextCommand(manifestOption, summaryStats, indexFreshness, untrackedCount, newlyExcludedCount);
     return {
         manifestSchema: manifest.schemaVersion,
-        trackedFileCount: Object.keys(manifest.files).length,
+        scannedFileCount: scannedFiles.length,
+        trackedFileCount: manifestFilePaths.size,
+        untrackedCount,
+        newlyExcludedCount,
         summaryFileCount: summaryStats.summaryFileCount,
         orphanCount: summaryStats.orphanCount,
         summaryFreshness: {
@@ -207,18 +237,23 @@ function formatIndexFreshness(freshness) {
 async function runStatus(options, config) {
     const manifestPath = (0, paths_1.resolveDiffdocArtifactPath)(options.manifest, config.baseDir);
     const manifest = await readManifest(manifestPath);
+    const commandCwd = process.cwd();
+    const repoPath = node_path_1.default.resolve(commandCwd, options.path || config.repoPath);
+    const scannedFiles = await scanEligibleFiles(repoPath, config);
     const summaryStats = await getSummaryStats(manifestPath, manifest);
     const indexFreshness = await getIndexFreshness(manifest, config);
-    const report = buildStatusReport(manifest, summaryStats, indexFreshness, options.manifest);
+    const report = buildStatusReport(manifest, scannedFiles, summaryStats, indexFreshness, options.manifest);
     if (options.json) {
         console.log(JSON.stringify(report, null, 2));
         return;
     }
     console.log(`manifest schema: ${report.manifestSchema}`);
-    console.log(`tracked files: ${report.trackedFileCount}`);
+    console.log(`scanned files: ${report.scannedFileCount}`);
+    console.log(`manifest files: ${report.trackedFileCount}`);
+    console.log(`untracked: ${report.untrackedCount}`);
+    console.log(`newly excluded: ${report.newlyExcludedCount}`);
     console.log(`summary files: ${report.summaryFileCount}`);
     console.log(`orphans: ${report.orphanCount}`);
-    console.log(`stale summaries: ${report.summaryFreshness.stale}`);
     console.log(`summary freshness: ${formatSummaryFreshness(summaryStats)}`);
     console.log(`index freshness: ${formatIndexFreshness(indexFreshness)}`);
     console.log("");

package/dist/commands/summarize.js

@@ -6,74 +6,21 @@ Object.defineProperty(exports, "__esModule", { value: true });
 exports.runSummarize = runSummarize;
 const promises_1 = __importDefault(require("node:fs/promises"));
 const node_path_1 = __importDefault(require("node:path"));
-const ignore_1 = __importDefault(require("ignore"));
 const artifacts_1 = require("../types/artifacts");
 const schemas_1 = require("../schemas");
 const git_1 = require("../utils/git");
 const hashing_1 = require("../utils/hashing");
 const llm_1 = require("../utils/llm");
 const paths_1 = require("../utils/paths");
+const scan_1 = require("../utils/scan");
 const MANIFEST_SCHEMA_URL = `${schemas_1.SCHEMA_BASE_URL}/manifest.schema.json`;
 const SUMMARY_ASSET_SCHEMA_URL = `${schemas_1.SCHEMA_BASE_URL}/summary-asset.schema.json`;
-function normalizeRelativePath(filePath) {
-    return filePath.split(node_path_1.default.sep).join("/");
-}
 function getSummaryDir(manifestPath) {
     return node_path_1.default.resolve(node_path_1.default.dirname(manifestPath), "summaries");
 }
 function getSummaryPath(summaryDir, hash) {
     return node_path_1.default.resolve(summaryDir, `${hash}.json`);
 }
-function normalizeGlobPattern(pattern) {
-    return pattern.split(node_path_1.default.sep).join("/");
-}
-function escapeRegex(value) {
-    return value.replace(/[|\\{}()[\]^$+?.]/g, "\\$&");
-}
-function globToRegExp(pattern) {
-    const normalized = normalizeGlobPattern(pattern);
-    let regexBody = "";
-    for (let i = 0; i < normalized.length; i += 1) {
-        const char = normalized[i];
-        const next = normalized[i + 1];
-        if (char === "*" && next === "*") {
-            regexBody += ".*";
-            i += 1;
-            continue;
-        }
-        if (char === "*") {
-            regexBody += "[^/]*";
-            continue;
-        }
-        if (char === "?") {
-            regexBody += "[^/]";
-            continue;
-        }
-        regexBody += escapeRegex(char);
-    }
-    return new RegExp(`^${regexBody}$`);
-}
-function compileGlobs(patterns) {
-    return patterns.filter(Boolean).map(globToRegExp);
-}
-function matchesAny(filePath, patterns) {
-    return patterns.some((pattern) => pattern.test(filePath));
-}
-function shouldIncludeFile(filePath, includeGlobs, excludeGlobs, ignoreMatcher) {
-    if (ignoreMatcher.ignores(filePath)) {
-        return false;
-    }
-    if (excludeGlobs.length > 0 && matchesAny(filePath, excludeGlobs)) {
-        return false;
-    }
-    if (includeGlobs.length > 0 && !matchesAny(filePath, includeGlobs)) {
-        return false;
-    }
-    return true;
-}
-function isIgnoredDirectory(dirPath, ignoreMatcher) {
-    return ignoreMatcher.ignores(dirPath) || ignoreMatcher.ignores(`${dirPath}/`);
-}
 async function atomicWriteUtf8(targetPath, content) {
     await promises_1.default.mkdir(node_path_1.default.dirname(targetPath), { recursive: true });
     const tempPath = `${targetPath}.${process.pid}.${Date.now()}.tmp`;
@@ -182,44 +129,6 @@ async function readManifest(manifestPath) {
         throw error;
     }
 }
-async function readIgnoreMatcher(repoPath, ignoreFilePath) {
-    const matcher = (0, ignore_1.default)();
-    const absolutePath = node_path_1.default.isAbsolute(ignoreFilePath)
-        ? ignoreFilePath
-        : node_path_1.default.resolve(repoPath, ignoreFilePath);
-    try {
-        const raw = await promises_1.default.readFile(absolutePath, "utf8");
-        return matcher.add(raw);
-    }
-    catch (error) {
-        const nodeError = error;
-        if (nodeError.code === "ENOENT") {
-            return matcher;
-        }
-        throw error;
-    }
-}
-async function walkCodeFiles(rootPath, includeGlobs, excludeGlobs, ignoreMatcher, currentPath = rootPath) {
-    const entries = await promises_1.default.readdir(currentPath, { withFileTypes: true });
-    const files = [];
-    for (const entry of entries) {
-        const entryPath = node_path_1.default.join(currentPath, entry.name);
-        if (entry.isDirectory()) {
-            const relativePath = normalizeRelativePath(node_path_1.default.relative(rootPath, entryPath));
-            if (!isIgnoredDirectory(relativePath, ignoreMatcher)) {
-                files.push(...await walkCodeFiles(rootPath, includeGlobs, excludeGlobs, ignoreMatcher, entryPath));
-            }
-            continue;
-        }
-        if (entry.isFile()) {
-            const relativePath = normalizeRelativePath(node_path_1.default.relative(rootPath, entryPath));
-            if (shouldIncludeFile(relativePath, includeGlobs, excludeGlobs, ignoreMatcher)) {
-                files.push(relativePath);
-            }
-        }
-    }
-    return files.sort();
-}
 function countHashRefs(files) {
     const refs = new Map();
     for (const hash of Object.values(files)) {
@@ -330,19 +239,19 @@ async function runSummarize(options, config) {
     }
     const startedAt = new Date();
     const commandCwd = process.cwd();
-    const repoPath = node_path_1.default.resolve(commandCwd, options.path);
+    const repoPath = node_path_1.default.resolve(commandCwd, options.path || config.repoPath);
     const manifestPath = (0, paths_1.resolveDiffdocArtifactPath)(options.out, config.baseDir);
     const summaryDir = getSummaryDir(manifestPath);
     const manifest = await readManifest(manifestPath);
     const refs = countHashRefs(manifest.files);
-    const includePatterns = compileGlobs((options.includeGlobs && options.includeGlobs.length > 0)
-        ? options.includeGlobs.map(normalizeGlobPattern)
-        : config.summarize.includeGlobs.map(normalizeGlobPattern));
-    const excludePatterns = compileGlobs((options.excludeGlobs && options.excludeGlobs.length > 0)
-        ? options.excludeGlobs.map(normalizeGlobPattern)
-        : config.summarize.excludeGlobs.map(normalizeGlobPattern));
+    const includePatterns = (0, scan_1.compileGlobs)((options.includeGlobs && options.includeGlobs.length > 0)
+        ? options.includeGlobs.map(scan_1.normalizeGlobPattern)
+        : config.summarize.includeGlobs.map(scan_1.normalizeGlobPattern));
+    const excludePatterns = (0, scan_1.compileGlobs)((options.excludeGlobs && options.excludeGlobs.length > 0)
+        ? options.excludeGlobs.map(scan_1.normalizeGlobPattern)
+        : config.summarize.excludeGlobs.map(scan_1.normalizeGlobPattern));
     const ignoreFile = options.ignoreFile || config.summarize.ignoreFile;
-    const ignoreMatcher = await readIgnoreMatcher(repoPath, ignoreFile);
+    const ignoreMatcher = await (0, scan_1.readIgnoreMatcher)(repoPath, ignoreFile);
     const customPromptHash = getPromptHash(config);
     const customPromptSource = customPromptHash ? config.summarize.summaryPromptSource : undefined;
     const summaryFreshnessExpected = (hash) => ({
@@ -400,7 +309,7 @@ async function runSummarize(options, config) {
         manifest.files = {};
         refs.clear();
         await writeManifest(manifestPath, manifest);
-        const files = await walkCodeFiles(repoPath, includePatterns, excludePatterns, ignoreMatcher);
+        const files = await (0, scan_1.walkCodeFiles)(repoPath, includePatterns, excludePatterns, ignoreMatcher);
         const totalFiles = files.length;
         let completedFiles = 0;
         if (!isJson) {
@@ -462,7 +371,7 @@ async function runSummarize(options, config) {
                 totals.scanned += 1;
             });
             try {
-                if (!shouldIncludeFile(filePath, includePatterns, excludePatterns, ignoreMatcher)) {
+                if (!(0, scan_1.shouldIncludeFile)(filePath, includePatterns, excludePatterns, ignoreMatcher)) {
                     await withManifestLock(async () => {
                         const removed = await removeManifestPath(filePath, manifest, manifestPath, summaryDir, refs);
                         if (removed) {

package/dist/commands/validate.js

@@ -7,6 +7,7 @@ exports.runValidate = runValidate;
 const promises_1 = __importDefault(require("node:fs/promises"));
 const node_path_1 = __importDefault(require("node:path"));
 const artifacts_1 = require("../types/artifacts");
+const schemas_1 = require("../schemas");
 const paths_1 = require("../utils/paths");
 function getSummaryDir(manifestPath) {
     return node_path_1.default.resolve(node_path_1.default.dirname(manifestPath), "summaries");
@@ -80,6 +81,8 @@ async function runValidate(options, config) {
         }
     }
     const report = {
+        schemaVersion: schemas_1.SCHEMA_DIR_VERSION,
+        schemaBaseUri: schemas_1.SCHEMA_BASE_URL,
         valid: manifestValid && issues.length === 0,
         manifestPath,
         manifestValid,
@@ -91,6 +94,8 @@ async function runValidate(options, config) {
         console.log(JSON.stringify(report, null, 2));
     }
     else {
+        console.log(`Schema version: v${schemas_1.SCHEMA_DIR_VERSION}`);
+        console.log(`Schema URI: ${schemas_1.SCHEMA_BASE_URL}`);
         console.log(`Manifest: ${manifestPath}`);
         console.log(`Manifest valid: ${manifestValid ? "yes" : "NO"}`);
         console.log(`Summary assets checked: ${summaryAssetsChecked}`);

package/dist/config.js

@@ -123,6 +123,7 @@ function buildRuntimeConfig(options, needs = { chat: true, embeddings: true }) {
     }
     return {
         baseDir: readOption(mergedOptions.baseDir, "DIFFDOC_BASE_DIR", "./.diffdoc"),
+        repoPath: readOption(mergedOptions.repoPath, "DIFFDOC_REPO_PATH", "."),
         provider,
         chat: {
             apiKey,

package/dist/index.js

@@ -5,6 +5,7 @@ const commander_1 = require("commander");
 const config_1 = require("./config");
 const embed_1 = require("./commands/embed");
 const init_1 = require("./commands/init");
+const prune_1 = require("./commands/prune");
 const query_1 = require("./commands/query");
 const status_1 = require("./commands/status");
 const summarize_1 = require("./commands/summarize");
@@ -43,7 +44,7 @@ function addCloudEndpointAndKeyOptions(command) {
 program
     .name("diffdoc")
     .description("Translate repository code shifts into plain-English business context")
-    .version("0.6.0");
+    .version("0.6.4");
 program
     .command("init")
     .description("Initialize DiffDoc configuration for this repository")
@@ -146,12 +147,31 @@ addCloudEndpointAndKeyOptions(addEmbeddingOptions(addBaseOptions(program
 addBaseOptions(program
     .command("status"))
     .description("Show manifest and index sync status")
+    .option("--path <path>", "repository or code path to scan")
     .option("--manifest <path>", "manifest input path under --base-dir", "manifest.json")
     .option("--json", "print status as JSON for CI", false)
     .action(async (options) => {
     try {
         const config = (0, config_1.buildRuntimeConfig)(options, { embeddings: false, chat: false });
-        await (0, status_1.runStatus)({ manifest: options.manifest, json: options.json }, config);
+        await (0, status_1.runStatus)({ path: options.path, manifest: options.manifest, json: options.json }, config);
+    }
+    catch (error) {
+        console.error(error instanceof Error ? error.message : error);
+        process.exit(1);
+    }
+});
+addBaseOptions(program
+    .command("prune"))
+    .description("Remove manifest entries for deleted or excluded files")
+    .option("--path <path>", "repository or code path to scan")
+    .option("--manifest <path>", "manifest input path under --base-dir", "manifest.json")
+    .option("--dry-run", "show what would be pruned without executing", false)
+    .option("--yes", "skip confirmation prompt", false)
+    .option("--json", "print prune report as JSON for CI", false)
+    .action(async (options) => {
+    try {
+        const config = (0, config_1.buildRuntimeConfig)(options, { embeddings: false, chat: false });
+        await (0, prune_1.runPrune)({ path: options.path, manifest: options.manifest, dryRun: options.dryRun, yes: options.yes, json: options.json }, config);
     }
     catch (error) {
         console.error(error instanceof Error ? error.message : error);

package/dist/schemas.js

@@ -15,6 +15,7 @@ exports.SCHEMA_BASE_URL = `https://raw.githubusercontent.com/sullyTheDev/diffdoc
 exports.DiffdocConfigSchema = zod_1.z.object({
     $schema: zod_1.z.string().optional(),
     baseDir: zod_1.z.string().optional(),
+    repoPath: zod_1.z.string().optional(),
     aiProvider: zod_1.z.enum(["local", "cloud"]).optional(),
     localLlmEndpoint: zod_1.z.string().optional(),
     localEmbedEndpoint: zod_1.z.string().optional(),

package/dist/utils/scan.js

@@ -0,0 +1,105 @@
+"use strict";
+var __importDefault = (this && this.__importDefault) || function (mod) {
+    return (mod && mod.__esModule) ? mod : { "default": mod };
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.normalizeRelativePath = normalizeRelativePath;
+exports.normalizeGlobPattern = normalizeGlobPattern;
+exports.compileGlobs = compileGlobs;
+exports.shouldIncludeFile = shouldIncludeFile;
+exports.readIgnoreMatcher = readIgnoreMatcher;
+exports.walkCodeFiles = walkCodeFiles;
+const promises_1 = __importDefault(require("node:fs/promises"));
+const node_path_1 = __importDefault(require("node:path"));
+const ignore_1 = __importDefault(require("ignore"));
+function normalizeRelativePath(filePath) {
+    return filePath.split(node_path_1.default.sep).join("/");
+}
+function normalizeGlobPattern(pattern) {
+    return pattern.split(node_path_1.default.sep).join("/");
+}
+function escapeRegex(value) {
+    return value.replace(/[|\\{}()[\]^$+?.]/g, "\\$&");
+}
+function globToRegExp(pattern) {
+    const normalized = normalizeGlobPattern(pattern);
+    let regexBody = "";
+    for (let i = 0; i < normalized.length; i += 1) {
+        const char = normalized[i];
+        const next = normalized[i + 1];
+        if (char === "*" && next === "*") {
+            regexBody += ".*";
+            i += 1;
+            continue;
+        }
+        if (char === "*") {
+            regexBody += "[^/]*";
+            continue;
+        }
+        if (char === "?") {
+            regexBody += "[^/]";
+            continue;
+        }
+        regexBody += escapeRegex(char);
+    }
+    return new RegExp(`^${regexBody}$`);
+}
+function compileGlobs(patterns) {
+    return patterns.filter(Boolean).map(globToRegExp);
+}
+function matchesAny(filePath, patterns) {
+    return patterns.some((pattern) => pattern.test(filePath));
+}
+function shouldIncludeFile(filePath, includeGlobs, excludeGlobs, ignoreMatcher) {
+    if (ignoreMatcher.ignores(filePath)) {
+        return false;
+    }
+    if (excludeGlobs.length > 0 && matchesAny(filePath, excludeGlobs)) {
+        return false;
+    }
+    if (includeGlobs.length > 0 && !matchesAny(filePath, includeGlobs)) {
+        return false;
+    }
+    return true;
+}
+function isIgnoredDirectory(dirPath, ignoreMatcher) {
+    return ignoreMatcher.ignores(dirPath) || ignoreMatcher.ignores(`${dirPath}/`);
+}
+async function readIgnoreMatcher(repoPath, ignoreFilePath) {
+    const matcher = (0, ignore_1.default)();
+    const absolutePath = node_path_1.default.isAbsolute(ignoreFilePath)
+        ? ignoreFilePath
+        : node_path_1.default.resolve(repoPath, ignoreFilePath);
+    try {
+        const raw = await promises_1.default.readFile(absolutePath, "utf8");
+        return matcher.add(raw);
+    }
+    catch (error) {
+        const nodeError = error;
+        if (nodeError.code === "ENOENT") {
+            return matcher;
+        }
+        throw error;
+    }
+}
+async function walkCodeFiles(rootPath, includeGlobs, excludeGlobs, ignoreMatcher, currentPath = rootPath) {
+    const entries = await promises_1.default.readdir(currentPath, { withFileTypes: true });
+    const files = [];
+    for (const entry of entries) {
+        const entryPath = node_path_1.default.join(currentPath, entry.name);
+        if (entry.isDirectory()) {
+            const relativePath = normalizeRelativePath(node_path_1.default.relative(rootPath, entryPath));
+            if (!isIgnoredDirectory(relativePath, ignoreMatcher)) {
+                files.push(...await walkCodeFiles(rootPath, includeGlobs, excludeGlobs, ignoreMatcher, entryPath));
+            }
+            continue;
+        }
+        if (entry.isFile()) {
+            const relativePath = normalizeRelativePath(node_path_1.default.relative(rootPath, entryPath));
+            if (shouldIncludeFile(relativePath, includeGlobs, excludeGlobs, ignoreMatcher)) {
+                files.push(relativePath);
+            }
+        }
+    }
+    return files.sort();
+}

package/package.json

@@ -1,52 +1,52 @@
-{
-  "name": "diffdoc",
-  "version": "0.6.2",
-  "description": "Translate repository code shifts into plain-English business context",
-  "license": "MIT",
-  "author": "Christopher Sullivan",
-  "homepage": "https://github.com/sullyTheDev/diffdoc#readme",
-  "bugs": {
-    "url": "https://github.com/sullyTheDev/diffdoc/issues"
-  },
-  "repository": {
-    "type": "git",
-    "url": "git+https://github.com/sullyTheDev/diffdoc.git"
-  },
-  "type": "commonjs",
-  "main": "dist/index.js",
-  "bin": {
-    "diffdoc": "dist/index.js",
-    "diffdoc-mcp": "dist/mcp.js"
-  },
-  "files": [
-    "dist",
-    "schemas",
-    "README.md",
-    "LICENSE",
-    ".diffdocrc.example"
-  ],
-  "engines": {
-    "node": ">=22"
-  },
-  "scripts": {
-    "build": "tsc && node dist/scripts/generate-schemas.js",
-    "generate:schemas": "node dist/scripts/generate-schemas.js",
-    "clean": "node -e \"require('fs').rmSync('dist', { recursive: true, force: true })\"",
-    "start": "tsc && node ./dist/index.js",
-    "prepare": "npm run build"
-  },
-  "dependencies": {
-    "@modelcontextprotocol/sdk": "^1.29.0",
-    "commander": "^12.0.0",
-    "ignore": "^7.0.5",
-    "openai": "^4.28.0",
-    "simple-git": "^3.24.0",
-    "vectra": "^0.14.0",
-    "zod": "^3.25.76"
-  },
-  "devDependencies": {
-    "@types/node": "^20.19.41",
-    "typescript": "^5.3.3",
-    "zod-to-json-schema": "^3.25.2"
-  }
-}
+{
+  "name": "diffdoc",
+  "version": "0.6.4",
+  "description": "Translate repository code shifts into plain-English business context",
+  "license": "MIT",
+  "author": "Christopher Sullivan",
+  "homepage": "https://github.com/sullyTheDev/diffdoc#readme",
+  "bugs": {
+    "url": "https://github.com/sullyTheDev/diffdoc/issues"
+  },
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/sullyTheDev/diffdoc.git"
+  },
+  "type": "commonjs",
+  "main": "dist/index.js",
+  "bin": {
+    "diffdoc": "dist/index.js",
+    "diffdoc-mcp": "dist/mcp.js"
+  },
+  "files": [
+    "dist",
+    "schemas",
+    "README.md",
+    "LICENSE",
+    ".diffdocrc.example"
+  ],
+  "engines": {
+    "node": ">=22"
+  },
+  "scripts": {
+    "build": "tsc && node dist/scripts/generate-schemas.js",
+    "generate:schemas": "node dist/scripts/generate-schemas.js",
+    "clean": "node -e \"require('fs').rmSync('dist', { recursive: true, force: true })\"",
+    "start": "tsc && node ./dist/index.js",
+    "prepare": "npm run build"
+  },
+  "dependencies": {
+    "@modelcontextprotocol/sdk": "^1.29.0",
+    "commander": "^12.0.0",
+    "ignore": "^7.0.5",
+    "openai": "^4.28.0",
+    "simple-git": "^3.24.0",
+    "vectra": "^0.14.0",
+    "zod": "^3.25.76"
+  },
+  "devDependencies": {
+    "@types/node": "^20.19.41",
+    "typescript": "^5.3.3",
+    "zod-to-json-schema": "^3.25.2"
+  }
+}

package/schemas/v2/diffdocrc.schema.json

@@ -10,6 +10,9 @@
         "baseDir": {
           "type": "string"
         },
+        "repoPath": {
+          "type": "string"
+        },
         "aiProvider": {
           "type": "string",
           "enum": [