LLM Wiki is a cross-platform desktop application that turns your documents into an organized, interlinked knowledge base — automatically. Instead of traditional RAG (retrieve-and-answer from scratch every time), the LLM incrementally builds and maintains a persistent wiki from your sources。
This report presents the forensic synthetic code analysis of nashsu/llm_wiki, a TypeScript project with 16,917 GitHub stars. SynthScan v2.0 examined 150,252 lines of code across 435 source files, recording 430 pattern matches distributed across 11 syntactic categories. The overall adjusted score of 5.2 places this repository in the Low AI signal band.
The scanner applied 160+ deterministic lexical heuristics, multi-line block detectors, abstract syntax tree depth profilers, and a cross-file Jaccard similarity matrix to construct a statistically normalised synthetic code estimate. All matches are individually weighted by severity coefficient and contextual multiplier before summation, and the resulting headline score is temporally discounted to account for the repository's development history relative to the commercial emergence of large language model coding tooling (November 2022 onward).
This chart maps the temporal evolution of the adjusted synthetic code score across successive scan runs. An upward trajectory indicates ongoing incorporation of AI-generated code or expanding LLM-assisted scaffolding; a stable or declining trajectory may reflect active human refactoring, code removal, or the adoption of stricter authorship policies. The dashed secondary line (right axis) independently tracks total raw pattern hit count, which can diverge from the normalised score when codebase size changes significantly between scans.
Classifies detected patterns by their diagnostic confidence and structural impact. CRITICAL patterns (coefficient 10) represent definitive synthetic signatures — hallucinated imports, explicit LLM attribution metadata — virtually never produced by human authors. HIGH (5) indicates strong structural tells such as cross-file repetition or cross-linguistic idioms. MEDIUM (2) covers recognisable conversational padding and AI-specific vocabulary. LOW (1) captures subtle indicators like tautological comments and generic boilerplate that require density to carry independent signal.
This horizontal bar chart decomposes the repository's raw synthetic code score by top-level directory, allowing you to pinpoint precisely which modules or components carry the highest AI authorship density. Directories with disproportionately high scores relative to their size warrant targeted manual review: concentrated AI signatures often trace back to mass-generated configuration layers, auto-ported test suites, LLM-scaffolded boilerplate classes, or entire subsystems authored under heavy copilot assistance. Use this view to prioritise your human code-review effort.
The scanner identified 430 distinct pattern matches across 11 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a statistically significant AI authorship indicator. Expand any category row to inspect the individual file paths, line numbers, code snippets, and the lexical context (CODE, COMMENT, or STRING) in which each match was detected.
Reading the findings table: The Severity column indicates the diagnostic confidence level (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — comment-context matches receive a ×1.5 weight because LLMs systematically over-annotate. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier as dense clusters constitute far stronger evidence of synthetic authorship than isolated hits.
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | scripts/debug_ollama_tokens.py | 35 | # 1687-1693). maxContextSize is in CHARACTERS. ──────────────────────────── | COMMENT |
| MEDIUM | src-tauri/src/commands/extract_images.rs | 179 | // ── PDF (pdfium) ──────────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src-tauri/src/commands/extract_images.rs | 456 | // ── PPTX / DOCX (zip) ────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src-tauri/src/commands/extract_images.rs | 572 | // ── Helpers ───────────────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src-tauri/src/commands/extract_images.rs | 681 | // ── Extract-and-save: write to disk, skip the base64 round-trip ──────── | COMMENT |
| MEDIUM | src-tauri/src/commands/extract_images.rs | 993 | // ── Tauri command bindings ───────────────────────────────────────────── | COMMENT |
| MEDIUM | src-tauri/src/commands/extract_images.rs | 1067 | // ── Tests ────────────────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src-tauri/src/commands/vectorstore.rs | 349 | // ────────────────────────────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src-tauri/src/commands/vectorstore.rs | 377 | // ────────────────────────────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src-tauri/src/commands/vectorstore.rs | 918 | // ────────────────────────────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src-tauri/src/commands/vectorstore.rs | 926 | // ────────────────────────────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src-tauri/src/commands/fs.rs | 2481 | // ── collect_related_pages: regression coverage for the three match ───── | COMMENT |
| MEDIUM | src-tauri/src/commands/fs.rs | 2677 | // ── copy_directory: folder import recursion + filtering ────────── | COMMENT |
| MEDIUM | src/lib/lint.ts | 29 | // ── helpers ─────────────────────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/lint.ts | 108 | // ── Structural lint ─────────────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/lint.ts | 200 | // ── Semantic lint ───────────────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/enrich-wikilinks.real-llm.test.ts | 145 | // ── Scenario-specific strict assertions ──────────────────────── | COMMENT |
| MEDIUM | src/lib/sweep-reviews.scenarios.test.ts | 50 | // ── Setup ──────────────────────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/sweep-reviews.scenarios.test.ts | 69 | // ── Scenario runner ───────────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/sweep-reviews.scenarios.test.ts | 177 | // ── Test cases ────────────────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/ingest-parse.test.ts | 37 | // ── Happy paths ───────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/ingest-parse.test.ts | 128 | // ── H1: CRLF normalization ───────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/ingest-parse.test.ts | 163 | // ── H2: Stream truncation ────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/ingest-parse.test.ts | 223 | // ── H3: Marker whitespace / case variants ────────────────────────── | COMMENT |
| MEDIUM | src/lib/ingest-parse.test.ts | 276 | // ── H5: Literal markers inside fenced code blocks ────────────────── | COMMENT |
| MEDIUM | src/lib/ingest-parse.test.ts | 368 | // ── H6: Empty path ────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/ingest-parse.test.ts | 383 | // ── Path traversal guard (security) ───────────────────────────────── | COMMENT |
| MEDIUM⚡ | src/lib/greeting-detector.ts | 21 | // ── English ────────────────────────────────────────────────────── | COMMENT |
| MEDIUM⚡ | src/lib/greeting-detector.ts | 27 | // ── Chinese (简体 + 繁體) ────────────────────────────────────── | COMMENT |
| MEDIUM⚡ | src/lib/greeting-detector.ts | 35 | // ── Japanese ──────────────────────────────────────────────────── | COMMENT |
| MEDIUM⚡ | src/lib/greeting-detector.ts | 38 | // ── Korean ────────────────────────────────────────────────────── | COMMENT |
| MEDIUM⚡ | src/lib/greeting-detector.ts | 41 | // ── European casual openers ───────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.test.ts | 16 | // ── Tiny helpers ──────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.test.ts | 24 | // ── stripFrontmatter ──────────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.test.ts | 56 | // ── Empty / trivial inputs ───────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.test.ts | 80 | // ── Heading breadcrumb tracking ──────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.test.ts | 142 | // ── Code block preservation ──────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.test.ts | 189 | // ── Table preservation ───────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.test.ts | 227 | // ── Recursive split levels ───────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.test.ts | 270 | // ── Overlap semantics ────────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.test.ts | 318 | // ── Small-chunk merging ──────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.test.ts | 346 | // ── Section-aware chunking (headings cut chunks) ─────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.test.ts | 368 | // ── Character offsets ────────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.test.ts | 391 | // ── Indices and determinism ──────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.test.ts | 420 | // ── End-to-end: realistic wiki page ──────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.test.ts | 518 | // ── CRLF content body (Windows line endings) ─────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.test.ts | 539 | // ── Nested fences (4-tick wrapping 3-tick) ───────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.test.ts | 570 | // ── Tables without a separator row ───────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.test.ts | 595 | // ── Contract checks on the output shape ──────────────────────────── | COMMENT |
| MEDIUM | src/lib/ingest-queue.test.ts | 1246 | // ── cleanupWrittenFiles — file delete + LanceDB chunk cascade ────── | COMMENT |
| MEDIUM | src/lib/text-chunker.ts | 91 | // ── Public API ──────────────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.ts | 129 | // ── Frontmatter handling ────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.ts | 151 | // ── Section segmentation ───────────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.ts | 245 | // ── Section → chunks ───────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.ts | 285 | // ── Atom tokenization (splittable vs indivisible) ──────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.ts | 376 | // ── Splittable atom → pieces ───────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.ts | 497 | // ── Piece sizing: pack pieces into chunks ≤ maxChars ───────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.ts | 538 | // ── Small-chunk merge pass ─────────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/text-chunker.ts | 563 | // ── Overlap injection ──────────────────────────────────────────────────── | COMMENT |
| MEDIUM | src/lib/wiki-page-delete.test.ts | 131 | // ── Media cascade: source-summary deletion drops images too ────── | COMMENT |
| 107 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | scripts/debug_ollama_tokens.py | 36 | def app_generation_max_tokens(max_context_chars: int) -> int: | CODE |
| LOW | src/App.tsx | 98 | async function hydrateScheduledImportAfterOpen(proj: WikiProject): Promise<void> { | CODE |
| LOW | src/stores/wiki-store.ts | 251 | export function createDefaultGraphUiState(): GraphUiState { | CODE |
| LOW | src/components/lint/lint-view.tsx | 39 | export function groupLintResultsForDisplay(results: readonly LintItem[]): { | CODE |
| LOW | src/components/chat/chat-panel.tsx | 114 | async function readAgentActivitySnapshot(path: string): Promise<string | null> { | CODE |
| LOW | src/components/chat/chat-panel.tsx | 275 | function backendReferenceToMessageReference(ref: BackendAgentReference): MessageReference { | CODE |
| LOW | src/components/chat/chat-panel.tsx | 418 | function summarizeAgentStepsForResume(steps: ChatAgentStep[] = []): string { | CODE |
| LOW | src/components/chat/chat-panel.tsx | 430 | function compactChatHistoryForResume( | CODE |
| LOW | src/components/chat/chat-panel.tsx | 1625 | function GeneratedOutputPreviewDialog({ | CODE |
| LOW | src/components/chat/chat-panel.tsx | 1668 | function ChatReferencePreviewPanel({ | CODE |
| LOW | src/components/chat/chat-panel.tsx | 1952 | function ChatReferencePreviewContent({ preview }: { preview: ChatReferencePreview }) { | CODE |
| LOW | src/components/chat/chat-panel.tsx | 1966 | function clampReferencePreviewWidth(width: number): number { | CODE |
| LOW | src/components/chat/chat-panel.tsx | 1970 | function ChatMarkdownReferencePreview({ preview }: { preview: ChatReferencePreview }) { | CODE |
| LOW | src/components/chat/chat-message.tsx | 302 | function extractShellApprovalCommand(steps: ChatAgentStep[]): string | null { | CODE |
| LOW | src/components/layout/icon-sidebar.tsx | 68 | function handleResearchPanelToggle() { | CODE |
| LOW | src/components/layout/research-panel-nav.ts | 14 | export function nextResearchPanelNavState( | CODE |
| LOW | src/components/graph/graph-view.tsx | 140 | function createGraphNodeHoverRenderer(palette: GraphThemePalette): NodeHoverDrawingFunction { | CODE |
| LOW | src/components/project/create-project-dialog.tsx | 33 | export function getCreateProjectFormStatus( | CODE |
| LOW | src/components/sources/sources-view.tsx | 282 | async function handleOpenSourceExternally(node: FileNode) { | CODE |
| LOW | src/lib/web-search.ts | 110 | export function hasConfiguredSearchProvider(config: SearchApiConfig): boolean { | CODE |
| LOW | src/lib/web-search.ts | 119 | export function hasConfiguredDeepResearchSources(config: SearchApiConfig): boolean { | CODE |
| LOW | src/lib/ingest-cache.ts | 42 | export async function listIngestedSourceIdentities(projectPath: string): Promise<string[]> { | CODE |
| LOW | src/lib/graph-visibility.ts | 8 | export function shouldHideEdgeByNodeTypes( | CODE |
| LOW | src/lib/claude-cli-transport.ts | 31 | export function createClaudeCodeStreamParser() { | CODE |
| LOW | src/lib/ingest-sanitize.ts | 58 | export function sanitizeIngestedFileContent(content: string): string { | CODE |
| LOW | src/lib/ingest-sanitize.ts | 120 | function stripFrontmatterKeyPrefix(content: string): string { | CODE |
| LOW | src/lib/ingest-sanitize.ts | 126 | function addMissingOpeningFrontmatterFence(content: string): string { | CODE |
| LOW | src/lib/ingest-sanitize.ts | 155 | function repairWikilinkListsInFrontmatter(content: string): string { | CODE |
| LOW | src/lib/detect-language.ts | 53 | function detectArabicScriptLanguage(text: string): "Arabic" | "Persian" { | CODE |
| LOW | src/lib/selection-edit.ts | 6 | export function normalizeSelectionReplacement(content: string): string { | CODE |
| LOW | src/lib/selection-edit.ts | 17 | export function normalizeEditableMarkdown(content: string): string { | CODE |
| LOW | src/lib/file-types.ts | 182 | export function isExtractedTextPreviewFile(filePath: string): boolean { | CODE |
| LOW | src/lib/embedding.test.ts | 83 | async function fetchEmbeddingViaMockHttp( | CODE |
| LOW | src/lib/embedding.test.ts | 171 | function isGoogleEmbeddingConfigForTest(config: { endpoint: string }): boolean { | CODE |
| LOW⚡ | src/lib/embedding.test.ts | 190 | function isVolcengineEndpointForTest(endpoint: string): boolean { | CODE |
| LOW⚡ | src/lib/embedding.test.ts | 199 | function volcengineEndpointForTest(config: { endpoint: string; model: string }): string { | CODE |
| LOW⚡ | src/lib/embedding.test.ts | 205 | function appendEndpointPathForTest(endpoint: string, suffix: string): string { | CODE |
| LOW | src/lib/graph-insights.ts | 31 | export function findSurprisingConnections( | CODE |
| LOW | src/lib/review-create-page.ts | 34 | function extractMissingPageCandidates(text: string): string[] { | CODE |
| LOW | src/lib/project-file-sync.ts | 130 | function scheduleRefreshAfterFileChanges(tasks: FileChangeTask[]): void { | CODE |
| LOW | src/lib/project-file-sync.ts | 188 | async function migrateUnchangedSourceMoves( | CODE |
| LOW | src/lib/project-file-sync.ts | 316 | function isRawSourcePathForCascade(relativePath: string): boolean { | CODE |
| LOW | src/lib/reasoning-detector.ts | 42 | export function countReasoningCharsInLine(rawLine: string): number { | CODE |
| LOW | src/lib/reasoning-detector.ts | 50 | export function extractReasoningTextFromLine(rawLine: string): string[] { | CODE |
| LOW | src/lib/mineru.ts | 214 | function mineruExtensionForMimeType(mimeType: string): string | null { | CODE |
| LOW | src/lib/mineru.ts | 326 | function convertHtmlTablesInSegment(segment: string): string { | CODE |
| LOW | src/lib/mineru.ts | 358 | function convertHtmlTablesToMarkdown(markdown: string): string { | CODE |
| LOW | src/lib/mineru.ts | 380 | function rewriteMineruMarkdownImages(markdown: string, pathMap: Map<string, string>): string { | CODE |
| LOW | src/lib/mineru.ts | 622 | async function downloadAndExtractMarkdown( | CODE |
| LOW | src/lib/mineru.ts | 683 | function localMineruBackendForVersion( | CODE |
| LOW | src/lib/page-merge.ts | 212 | export function stripBodyWikilinkPathPrefixes(content: string): string { | CODE |
| LOW | src/lib/page-merge.ts | 224 | function normalizeWikilinksOutsideCode(body: string): string { | CODE |
| LOW | src/lib/source-identity.ts | 39 | export function sourceSummarySlugFromIdentity(sourceIdentity: string): string { | CODE |
| LOW | src/lib/source-identity.ts | 65 | export function legacySourceSummarySlugFromIdentity(sourceIdentity: string): string { | CODE |
| LOW | src/lib/source-identity.ts | 84 | export function sourceSummarySlugCandidatesFromIdentity(sourceIdentity: string): string[] { | CODE |
| LOW | src/lib/source-identity.ts | 91 | function previousReadableSourceSummarySlugFromIdentity(sourceIdentity: string): string { | CODE |
| LOW | src/lib/llm-providers.ts | 155 | export function isLocalOrPrivateHttpEndpoint(endpoint: string): boolean { | CODE |
| LOW | src/lib/llm-providers.ts | 339 | function stripWireAgnosticOverrides(overrides?: RequestOverrides): Omit<RequestOverrides, "reasoning"> { | CODE |
| LOW | src/lib/llm-providers.ts | 355 | function supportsDeepSeekThinkingParam(config: LlmConfig): boolean { | CODE |
| LOW | src/lib/llm-providers.ts | 380 | function isOpenAiStrictCompletionModel(config: LlmConfig): boolean { | CODE |
| 110 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | .github/workflows/build.yml | 61 | if: startsWith(matrix.platform, 'ubuntu-22.04') | COMMENT |
| LOW | src-tauri/Cargo.toml | 41 | office_oxide = "=0.1.2" | COMMENT |
| LOW | src-tauri/src/proxy.rs | 1 | //! Global outbound HTTP proxy plumbing. | COMMENT |
| LOW | src-tauri/src/proxy.rs | 61 | /// Read `proxyConfig` out of the project's `app-state.json`. Returns | COMMENT |
| LOW | src-tauri/src/proxy.rs | 81 | /// instant. Two callers exist: (1) the Tauri setup hook, which | COMMENT |
| LOW | src-tauri/src/lib.rs | 561 | None::<Vec<&str>>, | COMMENT |
| LOW | src-tauri/src/agent/runtime.rs | 81 | // only when the user explicitly asks to update/overwrite an existing page; | COMMENT |
| LOW | src-tauri/src/agent/types.rs | 101 | pub tools: AgentToolOptions, | COMMENT |
| LOW | src-tauri/src/agent/types.rs | 121 | #[serde(default)] | COMMENT |
| LOW | src-tauri/src/agent/types.rs | 181 | pub snippet: Option<String>, | COMMENT |
| LOW | src-tauri/src/agent/provider.rs | 41 | #[serde(default)] | COMMENT |
| LOW | src-tauri/src/commands/extract_images.rs | 1 | //! Image extraction from PDF / PPTX / DOCX (Phase 1 of the multimodal | COMMENT |
| LOW | src-tauri/src/commands/extract_images.rs | 41 | /// (each image base64'd is ~MB-scale) AND blow up downstream VLM | COMMENT |
| LOW | src-tauri/src/commands/extract_images.rs | 81 | pub index: u32, | COMMENT |
| LOW | src-tauri/src/commands/extract_images.rs | 181 | /// Combined PDF text + image extraction in a single pdfium session. | COMMENT |
| LOW | src-tauri/src/commands/extract_images.rs | 681 | // ── Extract-and-save: write to disk, skip the base64 round-trip ──────── | COMMENT |
| LOW | src-tauri/src/commands/vectorstore.rs | 341 | .await | COMMENT |
| LOW | src-tauri/src/commands/vectorstore.rs | 361 | // chunk_index UInt32 0-based position within the page | COMMENT |
| LOW | src-tauri/src/commands/vectorstore.rs | 921 | // These exercise the pure data-shape and upsert/search/delete contracts | COMMENT |
| LOW | src-tauri/src/commands/claude_cli.rs | 1 | //! Claude Code CLI subprocess transport. | COMMENT |
| LOW | src-tauri/src/commands/fs_anydoc_benchmark.rs | 1 | //! Local-only comparison harness for AnyDoc and the compatibility parsers. | COMMENT |
| LOW | src-tauri/src/commands/fs.rs | 361 | } | COMMENT |
| LOW | src-tauri/src/commands/fs.rs | 381 | /// | COMMENT |
| LOW | src-tauri/src/commands/fs.rs | 401 | /// resource-dir logic. | COMMENT |
| LOW | src-tauri/src/commands/fs.rs | 421 | if let Ok(p) = std::env::var("PDFIUM_DYNAMIC_LIB_PATH") { | COMMENT |
| LOW | src-tauri/src/commands/fs.rs | 1841 | // | COMMENT |
| LOW | src-tauri/src/commands/fs.rs | 1921 | /// | COMMENT |
| LOW | src-tauri/src/commands/fs.rs | 2481 | // ── collect_related_pages: regression coverage for the three match ───── | COMMENT |
| LOW | src-tauri/src/commands/fs.rs | 2681 | // 1. Recursion goes ALL the way down (no depth cap) — users | COMMENT |
| LOW | src/components/layout/update-banner.tsx | 61 | dismissedVersion: result.remote, | COMMENT |
| LOW | src/components/search/search-view.tsx | 61 | }, | COMMENT |
| LOW | src/lib/ingest-sanitize.ts | 61 | // (1) Strip a code fence wrapping the whole document or just its | COMMENT |
| LOW | src/lib/markdown-image-resolver.ts | 121 | COMMENT | |
| LOW | src/lib/markdown-image-resolver.ts | 141 | // embedded images as `media/<source-slug>/img-N.png` — a path | COMMENT |
| LOW | src/lib/wiki-page-delete.test.ts | 121 | it("skips removePageEmbedding when slug derivation yields empty (defensive)", async () => { | COMMENT |
| LOW | src/lib/llm-providers.ts | 681 | return ( | COMMENT |
| LOW | src/lib/llm-client.ts | 341 | } | COMMENT |
| LOW | src/lib/wiki-page-delete.ts | 81 | // embedding cascade) so a failure in either of those — already | COMMENT |
| LOW | src/lib/frontmatter.ts | 21 | // step before falling back to the locator below. | COMMENT |
| LOW | src/lib/embedding.real-llm.test.ts | 221 | "fetchEmbedding returned null on oversize input but error isn't the oversize-specific message — looksLikeOversiz | COMMENT |
| LOW | src/lib/embedding.real-llm.test.ts | 381 | } | COMMENT |
| LOW | src/lib/ingest.ts | 781 | ? mineruSavedImages | COMMENT |
| LOW | src/lib/ingest.ts | 861 | }) | COMMENT |
| LOW | src/lib/ingest.ts | 901 | // Now that read_file's combined extraction has put `` | COMMENT |
| LOW | src/lib/ingest.ts | 921 | // sees it, AND skip the post-write safety-net injection further | COMMENT |
| LOW | src/lib/ingest.ts | 1901 | const { blocks, warnings: parseWarnings, truncatedPaths } = parseFileBlocks(text) | COMMENT |
| LOW | src/lib/ingest.ts | 2021 | // 1. Frontmatter array fields (sources, tags, related) | COMMENT |
| LOW | src/lib/ingest.ts | 3481 | getStore().addMessage("system", "No files were written. The LLM response did not contain valid FILE blocks.") | COMMENT |
| LOW | src/lib/__tests__/llm-providers.test.ts | 521 | expect(body.max_tokens).toBe(8192) | COMMENT |
| LOW | src/lib/__tests__/llm-providers.test.ts | 541 | // TCP connection rather than 403, which surfaces as a generic | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | src-tauri/src/agent/runtime.rs | 3240 | .get("placeholder") | CODE |
| LOW | src-tauri/src/agent/runtime.rs | 5008 | {"type":"text","id":"watermark","label":"Watermark","placeholder":"Optional"}, | CODE |
| LOW | src/i18n/zh.json | 65 | "placeholder": "解释、改写、缩短……", | CODE |
| LOW | src/i18n/zh.json | 181 | "placeholder": "https://example.com/article-one\nhttps://example.com/article-two", | CODE |
| LOW | src/i18n/zh.json | 261 | "placeholder": "向你的 Wiki 提问...", | CODE |
| LOW | src/i18n/zh.json | 375 | "placeholder": "搜索 Wiki 页面...", | CODE |
| LOW | src/i18n/ru.json | 65 | "placeholder": "Объяснить, переписать, сократить...", | CODE |
| LOW | src/i18n/ru.json | 181 | "placeholder": "https://example.com/article-one\nhttps://example.com/article-two", | CODE |
| LOW | src/i18n/ru.json | 261 | "placeholder": "Спросите о своей вики...", | CODE |
| LOW | src/i18n/ru.json | 375 | "placeholder": "Поиск по страницам вики...", | CODE |
| LOW | src/i18n/en.json | 65 | "placeholder": "Explain, rewrite, shorten...", | CODE |
| LOW | src/i18n/en.json | 181 | "placeholder": "https://example.com/article-one\nhttps://example.com/article-two", | CODE |
| LOW | src/i18n/en.json | 261 | "placeholder": "Ask about your wiki...", | CODE |
| LOW | src/i18n/en.json | 375 | "placeholder": "Search wiki pages...", | CODE |
| LOW | src/i18n/it.json | 65 | "placeholder": "Spiega, riscrivi, accorcia...", | CODE |
| LOW | src/i18n/it.json | 181 | "placeholder": "https://example.com/article-one\nhttps://example.com/article-two", | CODE |
| LOW | src/i18n/it.json | 261 | "placeholder": "Fai una domanda sul tuo wiki...", | CODE |
| LOW | src/i18n/it.json | 375 | "placeholder": "Cerca nelle pagine wiki...", | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | src/components/sources/sources-view.tsx | 304 | // Step 8: Refresh everything (UI side — must run with parent | COMMENT |
| LOW⚡ | src/lib/sources-merge.test.ts | 355 | // Step 2: ingest test2.md, which also covers attention. LLM emits | COMMENT |
| LOW⚡ | src/lib/sources-merge.test.ts | 364 | // Step 3: user later deletes test2.md. sources-view's deletion | COMMENT |
| LOW | src/lib/mineru.ts | 457 | // Step 1: Get upload URL | COMMENT |
| LOW | src/lib/mineru.ts | 477 | // Step 2: Upload file binary (convert base64 back to binary) | COMMENT |
| LOW | src/lib/deep-research.ts | 383 | // Step 1: gather research sources — use multiple queries if available, | COMMENT |
| LOW | src/lib/deep-research.ts | 409 | // Step 2: LLM synthesis | COMMENT |
| LOW | src/lib/deep-research.ts | 478 | // Step 3: Validate before writing. A successful stream can still contain | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | src/components/sources/sources-view.tsx | 304 | // Step 8: Refresh everything (UI side — must run with parent | COMMENT |
| LOW⚡ | src/lib/sources-merge.test.ts | 355 | // Step 2: ingest test2.md, which also covers attention. LLM emits | COMMENT |
| LOW⚡ | src/lib/sources-merge.test.ts | 364 | // Step 3: user later deletes test2.md. sources-view's deletion | COMMENT |
| LOW | src/lib/mineru.ts | 457 | // Step 1: Get upload URL | COMMENT |
| LOW | src/lib/mineru.ts | 477 | // Step 2: Upload file binary (convert base64 back to binary) | COMMENT |
| LOW | src/lib/deep-research.ts | 383 | // Step 1: gather research sources — use multiple queries if available, | COMMENT |
| LOW | src/lib/deep-research.ts | 409 | // Step 2: LLM synthesis | COMMENT |
| LOW | src/lib/deep-research.ts | 478 | // Step 3: Validate before writing. A successful stream can still contain | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | plans/multimodal-images.md | 240 | ### Phase 4: Settings toggle + cost guardrails | COMMENT |
| MEDIUM | src/components/settings/llm-presets.ts | 273 | // NVIDIA's own reasoning / agentic models | COMMENT |
| MEDIUM | src/components/settings/llm-presets.ts | 281 | // Popular third-party agentic / open-weight | COMMENT |
| MEDIUM | src/lib/lint.test.ts | 260 | makeFileNode("rag.md", "# RAG\nRetrieval augmented generation uses vector search."), | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | package-lock.json | 9567 | "resolved": "https://registry.npmjs.org/robust-predicates/-/robust-predicates-3.0.3.tgz", | CODE |
| MEDIUM | src/components/layout/update-banner.tsx | 40 | // robust to (a) a newer release shipping between notification | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | src/lib/dedup.test.ts | 88 | 'Sure, here are the duplicates I found:\n\n{"groups": [{"slugs": ["foo","bar"], "reason": "synonyms", "confidence" | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | scripts/debug_ollama_tokens.py | 55 | except Exception as e: # noqa: BLE001 — we want every failure mode visible | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | scripts/debug_ollama_tokens.py | 31 | CODE |