Skip oversized discovered files before scoring

Author: jahooma

sdk/src/__tests__/initial-session-state.test.ts

@@ -121,16 +121,18 @@ describe('Initial Session State', () => {
         return ['src', '.git', 'knowledge.md', 'README.md', '.gitignore']
       }
       if (dirPath === '/test-project/src') {
-        return ['index.ts', 'utils.ts']
+        return ['index.ts', 'utils.ts', 'generated.ts']
       }
       return []
     }) as CodebuffFileSystem['readdir']
-    mockFs.stat = (async (filePath: string): Promise<MockStatResult> => ({
-      isDirectory: () =>
-        filePath === '/test-project/src' || filePath === '/test-project/.git',
-      isFile: () =>
-        filePath !== '/test-project/src' && filePath !== '/test-project/.git',
-    })) as CodebuffFileSystem['stat']
+    mockFs.stat = (async (filePath: string) =>
+      ({
+        isDirectory: () =>
+          filePath === '/test-project/src' || filePath === '/test-project/.git',
+        isFile: () =>
+          filePath !== '/test-project/src' && filePath !== '/test-project/.git',
+        size: filePath.endsWith('generated.ts') ? 1_000_001 : 100,
+      }) as MockStatResult & { size: number }) as CodebuffFileSystem['stat']
 
     const readFilePaths: string[] = []
     const originalReadFile = mockFs.readFile
@@ -151,6 +153,9 @@ describe('Initial Session State', () => {
     expect(sessionState.mainAgentState.messageHistory).toEqual([])
     expect(readFilePaths.some((p) => p.endsWith('src/index.ts'))).toBe(true)
     expect(readFilePaths.some((p) => p.endsWith('src/utils.ts'))).toBe(true)
+    expect(readFilePaths.some((p) => p.endsWith('src/generated.ts'))).toBe(
+      false,
+    )
     expect(readFilePaths.some((p) => p.endsWith('README.md'))).toBe(false)
     expect(readFilePaths.some((p) => p.endsWith('knowledge.md'))).toBe(true)
   })

sdk/src/run-state.ts

@@ -141,6 +141,8 @@ type ProjectIndexInput = {
   readFile?: (filePath: string) => string | null | Promise<string | null>
 }
 
+const DEFAULT_SYMBOL_PARSE_FILE_BYTES = 1_000_000
+
 async function computeProjectIndex(params: ProjectIndexInput): Promise<{
   fileTree: FileTreeNode[]
   fileTokenScores: Record<string, any>
@@ -190,20 +192,53 @@ function getProjectIndexInput(params: {
       cwd,
       fileTree: discoveredProject.fileTree,
       filePaths: discoveredProject.filePaths.sort(),
-      readFile: (filePath: string) =>
-        fs.readFile(path.join(cwd, filePath), 'utf8').catch((error) => {
-          logger.debug?.(
-            { filePath, error: getErrorObject(error) },
-            'Failed to read discovered project file for symbol scoring',
-          )
-          return null
-        }),
+      readFile: createDiscoveredProjectReader({ cwd, fs, logger }),
     }
   }
 
   return undefined
 }
 
+function createDiscoveredProjectReader(params: {
+  cwd: string
+  fs: CodebuffFileSystem
+  logger: Logger
+}): (filePath: string) => Promise<string | null> {
+  const { cwd, fs, logger } = params
+  const maxBytes = getSymbolParseFileByteLimit()
+
+  return async (filePath: string) => {
+    const fullPath = path.join(cwd, filePath)
+    try {
+      const stats = await fs.stat(fullPath)
+      if (getFileSize(stats) > maxBytes) {
+        return null
+      }
+      return await fs.readFile(fullPath, 'utf8')
+    } catch (error) {
+      logger.debug?.(
+        { filePath, error: getErrorObject(error) },
+        'Failed to read discovered project file for symbol scoring',
+      )
+      return null
+    }
+  }
+}
+
+function getFileSize(stats: Awaited<ReturnType<CodebuffFileSystem['stat']>>) {
+  return typeof stats.size === 'number' ? stats.size : 0
+}
+
+function getSymbolParseFileByteLimit() {
+  const raw = process.env.CODEBUFF_MAX_PARSE_FILE_BYTES
+  if (!raw) return DEFAULT_SYMBOL_PARSE_FILE_BYTES
+
+  const parsed = Number.parseInt(raw, 10)
+  return Number.isFinite(parsed) && parsed > 0
+    ? parsed
+    : DEFAULT_SYMBOL_PARSE_FILE_BYTES
+}
+
 /**
  * Helper to convert ChildProcess to Promise with stdout/stderr
  */