Repository Analysis

nashsu/llm_wiki

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。

5.7 Low AI signal View on GitHub

Analysis Overview

This report presents the forensic synthetic code analysis of nashsu/llm_wiki, a TypeScript project with 14,530 GitHub stars. SynthScan v2.0 examined 130,727 lines of code across 389 source files, recording 390 pattern matches distributed across 11 syntactic categories. The overall adjusted score of 5.7 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).

5.7
Adjusted Score
5.7
Raw Score
100%
Time Factor
2026-07-11
Last Push
14.5K
Stars
TypeScript
Language
130.7K
Lines of Code
389
Files
390
Pattern Hits
2026-07-14
Scan Date
0.00
HC Hit Rate

What These Metrics Mean

Adjusted Score
Primary synthetic code indicator. Raw score normalised per 1,000 lines of code and multiplied by the temporal discount factor. This is the definitive comparative metric — use it to rank repositories by AI authorship density.
Raw Score
The unmodified sum of all severity-weighted, context-multiplied pattern match scores before temporal discounting. Reflects the absolute signal strength independent of when the repository was last active.
Time Factor
The temporal discount multiplier (0–100%) applied to the raw score. Repositories last updated before ChatGPT's launch (Nov 2022) receive a 5% factor. Full signal is only assigned to repositories active in the post-adoption era (Jan 2024+).
Pattern Hits
Total count of individual pattern matches across all files and categories. A high hit count with a low score may indicate a very large codebase with isolated AI snippets; a low count with a high score indicates dense, concentrated AI signatures.
HC Hit Rate
High+Critical pattern hits per file, averaged across the repository. This orthogonal signal catches repositories where a few files are densely packed with high-severity AI tells — a strong indicator even when the normalised score appears moderate due to codebase size.
Lines of Code / Files
Total lines and files analysed. The scanner examines 94 file extensions. These denominators are used to normalise the score, enabling fair comparison between repositories of vastly different sizes.

Score History

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.

Severity Breakdown

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.

CRITICAL 0HIGH 0MEDIUM 174LOW 216

Directory Score Breakdown

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.

Pattern Findings

The scanner identified 390 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.

Decorative Section Separators167 hits · 508 pts
SeverityFileLineSnippetContext
MEDIUMscripts/debug_ollama_tokens.py35# 1687-1693). maxContextSize is in CHARACTERS. ────────────────────────────COMMENT
MEDIUMsrc-tauri/src/commands/extract_images.rs179// ── PDF (pdfium) ────────────────────────────────────────────────────────COMMENT
MEDIUMsrc-tauri/src/commands/extract_images.rs456// ── PPTX / DOCX (zip) ──────────────────────────────────────────────────COMMENT
MEDIUMsrc-tauri/src/commands/extract_images.rs572// ── Helpers ─────────────────────────────────────────────────────────────COMMENT
MEDIUMsrc-tauri/src/commands/extract_images.rs681// ── Extract-and-save: write to disk, skip the base64 round-trip ────────COMMENT
MEDIUMsrc-tauri/src/commands/extract_images.rs993// ── Tauri command bindings ─────────────────────────────────────────────COMMENT
MEDIUMsrc-tauri/src/commands/extract_images.rs1067// ── Tests ──────────────────────────────────────────────────────────────COMMENT
MEDIUMsrc-tauri/src/commands/vectorstore.rs348// ──────────────────────────────────────────────────────────────────────────COMMENT
MEDIUMsrc-tauri/src/commands/vectorstore.rs376// ──────────────────────────────────────────────────────────────────────────COMMENT
MEDIUMsrc-tauri/src/commands/vectorstore.rs862// ──────────────────────────────────────────────────────────────────────────COMMENT
MEDIUMsrc-tauri/src/commands/vectorstore.rs870// ──────────────────────────────────────────────────────────────────────────COMMENT
MEDIUMsrc-tauri/src/commands/fs.rs2033 // ── collect_related_pages: regression coverage for the three match ─────COMMENT
MEDIUMsrc-tauri/src/commands/fs.rs2229 // ── copy_directory: folder import recursion + filtering ──────────COMMENT
MEDIUMsrc/lib/lint.ts28// ── helpers ───────────────────────────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/lint.ts148// ── Structural lint ───────────────────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/lint.ts300// ── Semantic lint ─────────────────────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/enrich-wikilinks.real-llm.test.ts145 // ── Scenario-specific strict assertions ────────────────────────COMMENT
MEDIUMsrc/lib/sweep-reviews.scenarios.test.ts50// ── Setup ────────────────────────────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/sweep-reviews.scenarios.test.ts69// ── Scenario runner ─────────────────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/sweep-reviews.scenarios.test.ts177// ── Test cases ──────────────────────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/ingest-parse.test.ts36// ── Happy paths ─────────────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/ingest-parse.test.ts154// ── H1: CRLF normalization ─────────────────────────────────────────COMMENT
MEDIUMsrc/lib/ingest-parse.test.ts189// ── H2: Stream truncation ──────────────────────────────────────────COMMENT
MEDIUMsrc/lib/ingest-parse.test.ts220// ── H3: Marker whitespace / case variants ──────────────────────────COMMENT
MEDIUMsrc/lib/ingest-parse.test.ts273// ── H5: Literal markers inside fenced code blocks ──────────────────COMMENT
MEDIUMsrc/lib/ingest-parse.test.ts365// ── H6: Empty path ──────────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/ingest-parse.test.ts380// ── Path traversal guard (security) ─────────────────────────────────COMMENT
MEDIUMsrc/lib/greeting-detector.ts21 // ── English ──────────────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/greeting-detector.ts27 // ── Chinese (简体 + 繁體) ──────────────────────────────────────COMMENT
MEDIUMsrc/lib/greeting-detector.ts35 // ── Japanese ────────────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/greeting-detector.ts38 // ── Korean ──────────────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/greeting-detector.ts41 // ── European casual openers ─────────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.test.ts16// ── Tiny helpers ────────────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.test.ts24// ── stripFrontmatter ────────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.test.ts56// ── Empty / trivial inputs ─────────────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.test.ts80// ── Heading breadcrumb tracking ────────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.test.ts142// ── Code block preservation ────────────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.test.ts189// ── Table preservation ─────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.test.ts227// ── Recursive split levels ─────────────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.test.ts270// ── Overlap semantics ──────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.test.ts318// ── Small-chunk merging ────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.test.ts346// ── Section-aware chunking (headings cut chunks) ───────────────────COMMENT
MEDIUMsrc/lib/text-chunker.test.ts368// ── Character offsets ──────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.test.ts391// ── Indices and determinism ────────────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.test.ts420// ── End-to-end: realistic wiki page ────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.test.ts518// ── CRLF content body (Windows line endings) ───────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.test.ts539// ── Nested fences (4-tick wrapping 3-tick) ─────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.test.ts570// ── Tables without a separator row ─────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.test.ts595// ── Contract checks on the output shape ────────────────────────────COMMENT
MEDIUMsrc/lib/ingest-queue.test.ts891// ── cleanupWrittenFiles — file delete + LanceDB chunk cascade ──────COMMENT
MEDIUMsrc/lib/text-chunker.ts86// ── Public API ────────────────────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.ts124// ── Frontmatter handling ──────────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.ts146// ── Section segmentation ─────────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.ts240// ── Section → chunks ─────────────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.ts280// ── Atom tokenization (splittable vs indivisible) ────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.ts371// ── Splittable atom → pieces ─────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.ts492// ── Piece sizing: pack pieces into chunks ≤ maxChars ─────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.ts533// ── Small-chunk merge pass ───────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/text-chunker.ts558// ── Overlap injection ────────────────────────────────────────────────────COMMENT
MEDIUMsrc/lib/wiki-page-delete.test.ts131 // ── Media cascade: source-summary deletion drops images too ──────COMMENT
107 more matches not shown…
Hyper-Verbose Identifiers147 hits · 139 pts
SeverityFileLineSnippetContext
LOWscripts/debug_ollama_tokens.py36def app_generation_max_tokens(max_context_chars: int) -> int:CODE
LOWsrc/App.tsx94 async function hydrateScheduledImportAfterOpen(proj: WikiProject): Promise<void> {CODE
LOWsrc/stores/wiki-store.ts228export function createDefaultGraphUiState(): GraphUiState {CODE
LOWsrc/components/lint/lint-view.tsx31export function groupLintResultsForDisplay(results: readonly LintItem[]): {CODE
LOWsrc/components/chat/chat-panel.tsx94async function readAgentActivitySnapshot(path: string): Promise<string | null> {CODE
LOWsrc/components/chat/chat-panel.tsx255function backendReferenceToMessageReference(ref: BackendAgentReference): MessageReference {CODE
LOWsrc/components/chat/chat-panel.tsx398function summarizeAgentStepsForResume(steps: ChatAgentStep[] = []): string {CODE
LOWsrc/components/chat/chat-panel.tsx410function compactChatHistoryForResume(CODE
LOWsrc/components/chat/chat-panel.tsx1575function GeneratedOutputPreviewDialog({CODE
LOWsrc/components/chat/chat-panel.tsx1618function ChatReferencePreviewPanel({CODE
LOWsrc/components/chat/chat-panel.tsx1703function ChatReferencePreviewContent({ preview }: { preview: ChatReferencePreview }) {CODE
LOWsrc/components/chat/chat-panel.tsx1717function clampReferencePreviewWidth(width: number): number {CODE
LOWsrc/components/chat/chat-panel.tsx1721function ChatMarkdownReferencePreview({ preview }: { preview: ChatReferencePreview }) {CODE
LOWsrc/components/chat/chat-message.tsx296function extractShellApprovalCommand(steps: ChatAgentStep[]): string | null {CODE
LOWsrc/components/layout/icon-sidebar.tsx68 function handleResearchPanelToggle() {CODE
LOWsrc/components/layout/research-panel-nav.ts14export function nextResearchPanelNavState(CODE
LOWsrc/components/graph/graph-view.tsx140function createGraphNodeHoverRenderer(palette: GraphThemePalette): NodeHoverDrawingFunction {CODE
LOWsrc/components/project/create-project-dialog.tsx33export function getCreateProjectFormStatus(CODE
LOWsrc/lib/web-search.ts108export function hasConfiguredSearchProvider(config: SearchApiConfig): boolean {CODE
LOWsrc/lib/web-search.ts117export function hasConfiguredDeepResearchSources(config: SearchApiConfig): boolean {CODE
LOWsrc/lib/graph-visibility.ts8export function shouldHideEdgeByNodeTypes(CODE
LOWsrc/lib/claude-cli-transport.ts31export function createClaudeCodeStreamParser() {CODE
LOWsrc/lib/ingest-sanitize.ts58export function sanitizeIngestedFileContent(content: string): string {CODE
LOWsrc/lib/ingest-sanitize.ts108function stripFrontmatterKeyPrefix(content: string): string {CODE
LOWsrc/lib/ingest-sanitize.ts114function addMissingOpeningFrontmatterFence(content: string): string {CODE
LOWsrc/lib/ingest-sanitize.ts143function repairWikilinkListsInFrontmatter(content: string): string {CODE
LOWsrc/lib/detect-language.ts53function detectArabicScriptLanguage(text: string): "Arabic" | "Persian" {CODE
LOWsrc/lib/selection-edit.ts6export function normalizeSelectionReplacement(content: string): string {CODE
LOWsrc/lib/selection-edit.ts17export function normalizeEditableMarkdown(content: string): string {CODE
LOWsrc/lib/file-types.ts156export function isExtractedTextPreviewFile(filePath: string): boolean {CODE
LOWsrc/lib/embedding.test.ts82async function fetchEmbeddingViaMockHttp(CODE
LOWsrc/lib/embedding.test.ts170function isGoogleEmbeddingConfigForTest(config: { endpoint: string }): boolean {CODE
LOWsrc/lib/embedding.test.ts189function isVolcengineEndpointForTest(endpoint: string): boolean {CODE
LOWsrc/lib/embedding.test.ts198function volcengineEndpointForTest(config: { endpoint: string; model: string }): string {CODE
LOWsrc/lib/embedding.test.ts204function appendEndpointPathForTest(endpoint: string, suffix: string): string {CODE
LOWsrc/lib/graph-insights.ts31export function findSurprisingConnections(CODE
LOWsrc/lib/review-create-page.ts34function extractMissingPageCandidates(text: string): string[] {CODE
LOWsrc/lib/project-file-sync.ts130function scheduleRefreshAfterFileChanges(tasks: FileChangeTask[]): void {CODE
LOWsrc/lib/project-file-sync.ts188async function migrateUnchangedSourceMoves(CODE
LOWsrc/lib/project-file-sync.ts316function isRawSourcePathForCascade(relativePath: string): boolean {CODE
LOWsrc/lib/reasoning-detector.ts42export function countReasoningCharsInLine(rawLine: string): number {CODE
LOWsrc/lib/reasoning-detector.ts50export function extractReasoningTextFromLine(rawLine: string): string[] {CODE
LOWsrc/lib/mineru.ts276function convertHtmlTablesInSegment(segment: string): string {CODE
LOWsrc/lib/mineru.ts308function convertHtmlTablesToMarkdown(markdown: string): string {CODE
LOWsrc/lib/mineru.ts330function rewriteMineruMarkdownImages(markdown: string, pathMap: Map<string, string>): string {CODE
LOWsrc/lib/mineru.ts572async function downloadAndExtractMarkdown(CODE
LOWsrc/lib/source-identity.ts39export function sourceSummarySlugFromIdentity(sourceIdentity: string): string {CODE
LOWsrc/lib/source-identity.ts65export function legacySourceSummarySlugFromIdentity(sourceIdentity: string): string {CODE
LOWsrc/lib/source-identity.ts84export function sourceSummarySlugCandidatesFromIdentity(sourceIdentity: string): string[] {CODE
LOWsrc/lib/source-identity.ts91function previousReadableSourceSummarySlugFromIdentity(sourceIdentity: string): string {CODE
LOWsrc/lib/llm-providers.ts121export function isLocalOrPrivateHttpEndpoint(endpoint: string): boolean {CODE
LOWsrc/lib/llm-providers.ts272function stripWireAgnosticOverrides(overrides?: RequestOverrides): Omit<RequestOverrides, "reasoning"> {CODE
LOWsrc/lib/llm-providers.ts285function supportsDeepSeekThinkingParam(config: LlmConfig): boolean {CODE
LOWsrc/lib/llm-providers.ts316function isOpenAiStrictCompletionModel(config: LlmConfig): boolean {CODE
LOWsrc/lib/llm-providers.ts330function adaptOpenAiStrictCompletionBody(config: LlmConfig, body: Record<string, unknown>): void {CODE
LOWsrc/lib/llm-providers.ts386function buildOpenAiCompatibleBody(CODE
LOWsrc/lib/llm-providers.ts544function buildAnthropicBodyWithReasoning(CODE
LOWsrc/lib/llm-providers.ts623function normalizeMiniMaxAnthropicBase(base: string): string {CODE
LOWsrc/lib/llm-providers.ts642function isOfficialMiniMaxAnthropicUrl(url: string): boolean {CODE
LOWsrc/lib/llm-providers.ts656function assertMiniMaxImageSupport(url: string, model: string, messages: ChatMessage[]): void {CODE
87 more matches not shown…
Over-Commented Block43 hits · 40 pts
SeverityFileLineSnippetContext
LOW.github/workflows/build.yml61 if: startsWith(matrix.platform, 'ubuntu-22.04')COMMENT
LOWsrc-tauri/Cargo.toml41arrow-schema = "57"COMMENT
LOWsrc-tauri/src/proxy.rs1//! Global outbound HTTP proxy plumbing.COMMENT
LOWsrc-tauri/src/proxy.rs21 "localhost,127.0.0.0/8,10.0.0.0/8,172.16.0.0/12,192.168.0.0/16,*.local";COMMENT
LOWsrc-tauri/src/proxy.rs61 serde_json::from_value(proxy.clone()).ok()COMMENT
LOWsrc-tauri/src/proxy.rs161mod tests {COMMENT
LOWsrc-tauri/src/lib.rs561 // Ark's api/coding/v3, etc.) still work. Requests leave the appCOMMENT
LOWsrc-tauri/src/agent/runtime.rs81 // only when the user explicitly asks to update/overwrite an existing page;COMMENT
LOWsrc-tauri/src/agent/types.rs101 #[serde(default)]COMMENT
LOWsrc-tauri/src/agent/types.rs181 #[serde(default, skip_serializing_if = "Option::is_none")]COMMENT
LOWsrc-tauri/src/commands/extract_images.rs1//! Image extraction from PDF / PPTX / DOCX (Phase 1 of the multimodalCOMMENT
LOWsrc-tauri/src/commands/extract_images.rs41 /// (each image base64'd is ~MB-scale) AND blow up downstream VLMCOMMENT
LOWsrc-tauri/src/commands/extract_images.rs81 pub index: u32,COMMENT
LOWsrc-tauri/src/commands/extract_images.rs181/// Combined PDF text + image extraction in a single pdfium session.COMMENT
LOWsrc-tauri/src/commands/extract_images.rs681// ── Extract-and-save: write to disk, skip the base64 round-trip ────────COMMENT
LOWsrc-tauri/src/commands/vectorstore.rs341 .map_err(|e| format!("Count error: {e}"))?;COMMENT
LOWsrc-tauri/src/commands/vectorstore.rs361// chunk_text Utf8 raw chunk content (for UI re-ranking + showingCOMMENT
LOWsrc-tauri/src/commands/vectorstore.rs861COMMENT
LOWsrc-tauri/src/commands/claude_cli.rs1//! Claude Code CLI subprocess transport.COMMENT
LOWsrc-tauri/src/commands/fs.rs181/// two threads simultaneously — interleaved calls are UB and haveCOMMENT
LOWsrc-tauri/src/commands/fs.rs221/// 2. Tauri resource dir (set via setup()) — the authoritative locationCOMMENT
LOWsrc-tauri/src/commands/fs.rs1601 // - test.md (unquoted, missed by Match 1)COMMENT
LOWsrc-tauri/src/commands/fs.rs1681/// without round-tripping them through the JS string-as-UTF8 pathCOMMENT
LOWsrc-tauri/src/commands/fs.rs1821 .unwrap()COMMENT
LOWsrc-tauri/src/commands/fs.rs2221 "concepts/rope.md",COMMENT
LOWsrc/components/settings/sections/about-section.tsx261 // system browser — `tauri-plugin-opener` is the official way toCOMMENT
LOWsrc/components/layout/update-banner.tsx61 dismissedVersion: result.remote,COMMENT
LOWsrc/components/search/search-view.tsx61 },COMMENT
LOWsrc/lib/ingest-sanitize.ts61 // (1) Strip an outer code fence wrapping the whole document.COMMENT
LOWsrc/lib/markdown-image-resolver.ts121COMMENT
LOWsrc/lib/markdown-image-resolver.ts141 // embedded images as `media/<source-slug>/img-N.png` — a pathCOMMENT
LOWsrc/lib/wiki-page-delete.test.ts121 it("skips removePageEmbedding when slug derivation yields empty (defensive)", async () => {COMMENT
LOWsrc/lib/wiki-page-delete.ts81 // embedding cascade) so a failure in either of those — alreadyCOMMENT
LOWsrc/lib/frontmatter.ts21// step before falling back to the locator below.COMMENT
LOWsrc/lib/embedding.real-llm.test.ts221 "fetchEmbedding returned null on oversize input but error isn't the oversize-specific message — looksLikeOversizCOMMENT
LOWsrc/lib/embedding.real-llm.test.ts381 }COMMENT
LOWsrc/lib/ingest.ts721 console.log(`[ingest:diag] cache-hit branch: starting image extraction for ${sp}`)COMMENT
LOWsrc/lib/ingest.ts801 // `wiki/media/<source-slug>/`. We DON'T inject the markdownCOMMENT
LOWsrc/lib/ingest.ts841 // Why this matters: an empty-alt image gets paraphrased away byCOMMENT
LOWsrc/lib/ingest.ts861 // `## Embedded Images` section to wiki/sources/<slug>.mdCOMMENT
LOWsrc/lib/ingest.ts1901 // wouldn't make semantic sense (they aren't source-derivedCOMMENT
LOWsrc/lib/__tests__/llm-providers.test.ts421 }) as Record<string, unknown>COMMENT
LOWsrc/lib/__tests__/llm-providers.test.ts441// "error sending request" (reqwest network error). Diagnosis:COMMENT
Verbosity Indicators8 hits · 14 pts
SeverityFileLineSnippetContext
LOWsrc/components/sources/sources-view.tsx187 // Step 8: Refresh everything (UI side — must run with parentCOMMENT
LOWsrc/lib/sources-merge.test.ts289 // Step 2: ingest test2.md, which also covers attention. LLM emitsCOMMENT
LOWsrc/lib/sources-merge.test.ts298 // Step 3: user later deletes test2.md. sources-view's deletionCOMMENT
LOWsrc/lib/mineru.ts407 // Step 1: Get upload URLCOMMENT
LOWsrc/lib/mineru.ts427 // Step 2: Upload file binary (convert base64 back to binary)COMMENT
LOWsrc/lib/deep-research.ts194 // Step 1: gather research sources — use multiple queries if available,COMMENT
LOWsrc/lib/deep-research.ts220 // Step 2: LLM synthesisCOMMENT
LOWsrc/lib/deep-research.ts289 // Step 3: Save to wikiCOMMENT
Structural Annotation Overuse8 hits · 14 pts
SeverityFileLineSnippetContext
LOWsrc/components/sources/sources-view.tsx187 // Step 8: Refresh everything (UI side — must run with parentCOMMENT
LOWsrc/lib/sources-merge.test.ts289 // Step 2: ingest test2.md, which also covers attention. LLM emitsCOMMENT
LOWsrc/lib/sources-merge.test.ts298 // Step 3: user later deletes test2.md. sources-view's deletionCOMMENT
LOWsrc/lib/mineru.ts407 // Step 1: Get upload URLCOMMENT
LOWsrc/lib/mineru.ts427 // Step 2: Upload file binary (convert base64 back to binary)COMMENT
LOWsrc/lib/deep-research.ts194 // Step 1: gather research sources — use multiple queries if available,COMMENT
LOWsrc/lib/deep-research.ts220 // Step 2: LLM synthesisCOMMENT
LOWsrc/lib/deep-research.ts289 // Step 3: Save to wikiCOMMENT
Modern AI Meta-Vocabulary4 hits · 11 pts
SeverityFileLineSnippetContext
MEDIUMplans/multimodal-images.md240### Phase 4: Settings toggle + cost guardrailsCOMMENT
MEDIUMsrc/components/settings/llm-presets.ts273 // NVIDIA's own reasoning / agentic modelsCOMMENT
MEDIUMsrc/components/settings/llm-presets.ts281 // Popular third-party agentic / open-weightCOMMENT
MEDIUMsrc/lib/lint.test.ts180 makeFileNode("rag.md", "# RAG\nRetrieval augmented generation uses vector search."),CODE
Fake / Example Data8 hits · 8 pts
SeverityFileLineSnippetContext
LOWsrc-tauri/src/agent/runtime.rs3152 .get("placeholder")CODE
LOWsrc-tauri/src/agent/runtime.rs4810 {"type":"text","id":"watermark","label":"Watermark","placeholder":"Optional"},CODE
LOWsrc/i18n/zh.json41 "placeholder": "解释、改写、缩短……",CODE
LOWsrc/i18n/zh.json194 "placeholder": "向你的 Wiki 提问...",CODE
LOWsrc/i18n/zh.json280 "placeholder": "搜索 Wiki 页面...",CODE
LOWsrc/i18n/en.json41 "placeholder": "Explain, rewrite, shorten...",CODE
LOWsrc/i18n/en.json194 "placeholder": "Ask about your wiki...",CODE
LOWsrc/i18n/en.json280 "placeholder": "Search wiki pages...",CODE
AI Slop Vocabulary2 hits · 5 pts
SeverityFileLineSnippetContext
MEDIUMpackage-lock.json9307 "resolved": "https://registry.npmjs.org/robust-predicates/-/robust-predicates-3.0.3.tgz",CODE
MEDIUMsrc/components/layout/update-banner.tsx40 // robust to (a) a newer release shipping between notificationCOMMENT
Slop Phrases1 hit · 2 pts
SeverityFileLineSnippetContext
MEDIUMsrc/lib/dedup.test.ts88 'Sure, here are the duplicates I found:\n\n{"groups": [{"slugs": ["foo","bar"], "reason": "synonyms", "confidence"CODE
Excessive Try-Catch Wrapping1 hit · 1 pts
SeverityFileLineSnippetContext
LOWscripts/debug_ollama_tokens.py55 except Exception as e: # noqa: BLE001 — we want every failure mode visibleCODE
Unused Imports1 hit · 1 pts
SeverityFileLineSnippetContext
LOWscripts/debug_ollama_tokens.py31CODE