Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Database info passed to research agent:', databaseInfo);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Requests targeting 127.0.0.1, localhost, or [::1] may access internal services not intended to be exposed.
return isDevelopment ? 'http://localhost:8000' : '';
Block requests to localhost and loopback addresses. Implement URL validation that rejects 127.x.x.x and ::1.
Tool names mimicking built-in system tools (e.g., 'bash', 'shell', 'terminal') can trick the LLM into routing actions to a malicious handler.
"/", StaticFiles(directory="ai-research-assistant/build", html=True), name="root"
Rename the tool to avoid colliding with system commands (bash, shell, exec, etc.).
API endpoints without rate limiting are vulnerable to brute force and denial of service.
@app.get("/")Add rate limiting middleware to all public API endpoints.
Using path.join or path.resolve with variables from user input without validation can lead to directory traversal.
output_file = os.path.join(args.output_dir, output_filename)
Sanitize user input before passing to path.join/resolve. Use path.normalize() and check for '..' sequences.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
result = await fresh_graph.ainvoke(initial_state, config=graph_config)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
String literals matching known API key prefixes (sk-, ghp_, AKIA, xoxb-, etc.) or long base64-like strings may expose secrets in source code.
OPENAI_API_KEY = os.environ.get("OPENAI_API_KEY", "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")Remove hardcoded secrets from source code. Use environment variables or a secrets manager.
subprocess calls with shell=True execute commands through the shell, enabling injection attacks.
result = subprocess.run(cmd, shell=True, capture_output=True, text=True, timeout=5) # Added 5 second timeout
Set shell=False in subprocess calls and pass command as a list.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
# Take a screenshot after navigating (using PDF as workaround)
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Detected SELECT * or ORM queries without explicit field selection. Returning all columns risks exposing sensitive fields (passwords, tokens, internal IDs) to the client or LLM context.
cursor.execute(f"SELECT * FROM {table} LIMIT 3")Always specify the exact columns or fields to return. Use SELECT with explicit column names or ORM select/projection options.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Stopping research from App.js');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('[DetailsPanel - renderWebLinks] Received links:', links);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('[DetailsPanel - renderWebLinks] Generated HTML for web links (first 300 chars):', generatedHtml.substring(0, 300) + '...');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('[DetailsPanel - generateEnhancedDetailsHtml] Processing item:', item);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Found enriched data:', item.enrichedData);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('[DetailsPanel - generateEnhancedDetailsHtml] item.enrichedData:', item.enrichedData);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Extracted links:', allLinks);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('[DetailsPanel - generateEnhancedDetailsHtml] Extracted links for renderWebLinks:', links);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Found node data:', item.nodeData);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('[DetailsPanel] Rendering todo plan - version in content:', versionMatch ? versionMatch[1] : 'unknown');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('[DetailsPanel] Parsed sections:', sections.length, sections.map(s => ({ title: s.title, taskCount: (s.content.match(/^[ ]*[-*] \[[ x~]/gm) || []).length })));Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('[DetailsPanel] Full todoPlanContent length:', todoPlanContent?.length);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Processing citation ${citationNumber}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Found references section:', referencesSection.substring(0, 200) + '...');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Found match with pattern:`, pattern, citationMatch);
Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Found URL in citation line:`, urlMatch[1]);
Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Searching entire document for citation ${citationNumber}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Found URL near citation:`, citationUrlMatch[1]);
Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Found URL in paragraph with citation:`, urlMatches[0]);
Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Using URL by position:`, url);
Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Falling back to first URL found:`, allUrls[0]);
Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`No URL found for citation ${citationNumber}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Error parsing URL for title extraction:', e);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('[RESEARCH_EVENT]', event.event_type || event.type, event)Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Node end: ${nodeName}`, event.data)Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log("Received research complete event:", event)Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Setting shouldAutoReconnect to ${value}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Manually disconnecting research event source');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Closing SSE connection manually');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Clearing stale connection timer');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('cancelResearch called');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Requesting backend to stop research for session ${currentSessionId}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Backend stop response:', data);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Error requesting backend stop:', err);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Canceling active reader');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Error canceling reader:', err);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Clearing connection check interval');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Successfully canceled research request for "${canceledRequest}"`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Clearing polling interval');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('[DEBUG] startResearch called with:', { query, provider, model, uploadedFileContent, databaseInfo });Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('[DEBUG] Stored event handlers:', { onEvent: typeof onEvent, onComplete: typeof onComplete, onError: typeof onError });Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Cleaning up any previous sessions for fresh start...');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('✅ Cleanup complete:', cleanupData);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('⚠️ Cleanup request failed (non-fatal):', cleanupErr);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
// console.log(`Starting research with URL: ${apiUrl}, steering: ${enableSteering}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Research settings: extraEffort=${extraEffort}, minimumEffort=${minimumEffort}, benchmarkMode=${benchmarkMode}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Ignoring duplicate request within debounce time');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Request already in progress, ignoring duplicate');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Cleaning up request: ${requestKey}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('[DEBUG] Adding database_info to request:', databaseInfo);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('[DEBUG] Converting uploaded file content array to string');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('[DEBUG] Added uploaded_data_content to request body, length:', combinedContent.length);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('[DEBUG] Added uploaded_data_content (string) to request body, length:', uploadedFileContent.length);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Making research request with body:', requestBody);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Research request response:', data);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
// console.log('[STEERING] Research started with steering enabled');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
// console.log(`[STEERING] Set current session ID: ${currentSessionId}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
// console.log('[STEERING] Starting plan polling for real-time updates');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Received stream URL: ${streamUrl}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`[DEBUG] connectToEventSource called with url: ${url}, onEvent handler:`, typeof onEvent);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Connecting to event source: ${fullUrl}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`SSE connection opened successfully! ReadyState: ${eventSource?.readyState}`, event);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Successfully reconnected to SSE stream.');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Research already complete - clearing stale connection timer');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Connection appears to be stale - no events in 30 seconds');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`[DEBUG] processEvent called with eventType: ${eventType}, event.data:`, event.data);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Received ${eventType} event with empty or undefined data`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`[DEBUG] Parsed data for ${eventType}:`, parsedData);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`[DEBUG] Received event type: ${eventType}`, parsedData);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Received array of ${parsedData.length} events for type ${eventType}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`[DEBUG] Calling onEvent for array item ${index + 1}/${parsedData.length}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`[DEBUG] Calling onEvent for single event of type ${eventType}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`[DEBUG] onEvent handler is null/undefined for event type ${eventType}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Research complete event received through named listener - disabling auto-reconnect');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Research complete - cleaning up request');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Closing SSE connection after receiving research_complete');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Clearing stale connection timer after research_complete');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`[DEBUG] Received event through 'message' listener:`, event.data);
Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Received generic message event with empty or undefined data`);
Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`[DEBUG] Received event through 'message' listener with type: ${eventType}`, data);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`[DEBUG] Calling onEvent from message listener for type: ${eventType}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`[DEBUG] onEvent handler is null/undefined in message listener`);
Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`[DEBUG] Received event through onmessage:`, event.data);
Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Received onmessage event with empty or undefined data`);
Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`[DEBUG] Parsed data in onmessage:`, parsedData);
Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Received array of ${parsedData.length} events via onmessage`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`[DEBUG] Skipping duplicate SSE event on fallback 'message' for: ${eventType}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Research complete event received in onmessage handler (array) - disabling auto-reconnect');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`[DEBUG] Processing event ${index + 1}/${parsedData.length} from onmessage array with type: ${eventType}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`[DEBUG] Calling onEvent from onmessage array for type: ${eventType}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`[DEBUG] Skipping duplicate SSE event on fallback 'message' for: ${eventType}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Research complete event received in onmessage handler (single) - disabling auto-reconnect');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Received unnamed single message event with type: ${eventType}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Ignoring EventSource error since research is already complete');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Auto-reconnect is enabled, will try to reconnect...');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Closing event source due to error and auto-reconnect disabled.');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Connection terminated without reconnect - cleaning up request');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
// console.log(`[STEERING] Starting polling for session: ${sessionId}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
// console.log(`[STEERING] Polling session status for: ${sessionId}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
// console.log('[STEERING] Session status:', sessionData);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Research completed, stopping polling');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
// console.log(`[STEERING] Sending message to session ${currentSessionId}: ${message}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
// console.log('[STEERING] Message sent successfully:', data);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
// console.log(`[STEERING] Fetching plan status for session: ${currentSessionId}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
// console.log('[STEERING] Received plan status:', {Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('[STEERING] Stopping plan polling');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`[STEERING] Getting todo plan for session: ${currentSessionId}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`[STEERING] Failed to get todo plan: ${response.status}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('[STEERING] Got todo plan:', planData);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Research is already complete - will not attempt to reconnect');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Not reconnecting as auto-reconnect is disabled');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Not reconnecting - cleaning up request');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Connection lost. Reconnection attempt ${attempt}/${maxAttempts} scheduled...`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Attempting reconnection ${attempt}/${maxAttempts}...`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Reconnecting to existing stream: ${lastResponseUrl}`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log(`Attempting to reconnect for query: "${lastQuery}"`);Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected console.log() usage in non-test source code. Console.log is not appropriate for production logging as it lacks log levels, structured output, and proper log management.
console.log('Maximum reconnection attempts reached.');Replace console.log with a structured logging library (e.g., winston, pino) that supports log levels and proper log management.
Detected patterns that disable, silence, or suppress logging or audit trails. Disabling security logging can mask malicious activity and hinder incident investigation.
console.log('Closing event source due to error and auto-reconnect disabled.');Ensure security-related logging is always enabled in production. Never suppress audit trails or security event logs.
Detected patterns that disable, silence, or suppress logging or audit trails. Disabling security logging can mask malicious activity and hinder incident investigation.
console.log('Not reconnecting as auto-reconnect is disabled');Ensure security-related logging is always enabled in production. Never suppress audit trails or security event logs.
Detected patterns that disable, silence, or suppress logging or audit trails. Disabling security logging can mask malicious activity and hinder incident investigation.
video.audio.write_audiofile(temp_audio_path, verbose=False, logger=None)
Ensure security-related logging is always enabled in production. Never suppress audit trails or security event logs.
Detected patterns that disable, silence, or suppress logging or audit trails. Disabling security logging can mask malicious activity and hinder incident investigation.
logger.info(f"callbacks at entry: {'present' if callbacks else 'none'}")Ensure security-related logging is always enabled in production. Never suppress audit trails or security event logs.
Detected patterns that disable, silence, or suppress logging or audit trails. Disabling security logging can mask malicious activity and hinder incident investigation.
# Silently fail - trajectory logging should never break research
Ensure security-related logging is always enabled in production. Never suppress audit trails or security event logs.
Detected patterns that disable, silence, or suppress logging or audit trails. Disabling security logging can mask malicious activity and hinder incident investigation.
logger.info(f"Tool class: {tool.__class__.__name__}, Config present: {hasattr(tool, 'config') and tool.config is not None}")Ensure security-related logging is always enabled in production. Never suppress audit trails or security event logs.
Detected patterns that disable, silence, or suppress logging or audit trails. Disabling security logging can mask malicious activity and hinder incident investigation.
logger.info(f"[SearchToolRegistry.__init__] Initializing registry with config type: {type(config).__name__ if config else 'None'}")Ensure security-related logging is always enabled in production. Never suppress audit trails or security event logs.
Detected patterns that disable, silence, or suppress logging or audit trails. Disabling security logging can mask malicious activity and hinder incident investigation.
self.logger.info("[VisualizationAgent] Defaulting to 'visualization_needed: False' due to parsing failure.")Ensure security-related logging is always enabled in production. Never suppress audit trails or security event logs.
Detected patterns that disable, silence, or suppress logging or audit trails. Disabling security logging can mask malicious activity and hinder incident investigation.
# else: # Old fallback logic based on regex content analysis (now handled by LLM or the final 'False' fallback)
Ensure security-related logging is always enabled in production. Never suppress audit trails or security event logs.
Detected patterns that disable, silence, or suppress logging or audit trails. Disabling security logging can mask malicious activity and hinder incident investigation.
self.logger.info(f"[VisualizationAgent] Final visualization needs keys: {vis_needs.keys() if vis_needs else 'None'}")Ensure security-related logging is always enabled in production. Never suppress audit trails or security event logs.
Detected patterns that disable, silence, or suppress logging or audit trails. Disabling security logging can mask malicious activity and hinder incident investigation.
def trace_state_changes(state, node_name=None):
Ensure security-related logging is always enabled in production. Never suppress audit trails or security event logs.
Requests targeting 127.0.0.1, localhost, or [::1] may access internal services not intended to be exposed.
const apiBaseUrl = isDevelopment ? 'http://localhost:8000' : ''
Block requests to localhost and loopback addresses. Implement URL validation that rejects 127.x.x.x and ::1.
Requests targeting 127.0.0.1, localhost, or [::1] may access internal services not intended to be exposed.
return isDevelopment ? 'http://localhost:8000' : ''
Block requests to localhost and loopback addresses. Implement URL validation that rejects 127.x.x.x and ::1.
Passing user-controlled variables directly to fetch, axios, or http.get without URL validation enables SSRF attacks.
const response = await fetch(apiUrl, {Validate and sanitize all URLs before making HTTP requests. Use an allowlist of permitted domains.
Requests targeting 127.0.0.1, localhost, or [::1] may access internal services not intended to be exposed.
base_url="http://localhost:3000"
Block requests to localhost and loopback addresses. Implement URL validation that rejects 127.x.x.x and ::1.
Requests targeting 127.0.0.1, localhost, or [::1] may access internal services not intended to be exposed.
base_url="http://localhost:3000"
Block requests to localhost and loopback addresses. Implement URL validation that rejects 127.x.x.x and ::1.
Requests targeting 127.0.0.1, localhost, or [::1] may access internal services not intended to be exposed.
base_url="http://localhost:3000"
Block requests to localhost and loopback addresses. Implement URL validation that rejects 127.x.x.x and ::1.
Requests targeting 127.0.0.1, localhost, or [::1] may access internal services not intended to be exposed.
# url="http://localhost:8931/sse" # Using SSE endpoint from docs
Block requests to localhost and loopback addresses. Implement URL validation that rejects 127.x.x.x and ::1.
Requests targeting 127.0.0.1, localhost, or [::1] may access internal services not intended to be exposed.
url = f"http://localhost:{server['default_port']}/mcp/v1/initialize"Block requests to localhost and loopback addresses. Implement URL validation that rejects 127.x.x.x and ::1.
Requests targeting 127.0.0.1, localhost, or [::1] may access internal services not intended to be exposed.
base_url="http://localhost:3000"
Block requests to localhost and loopback addresses. Implement URL validation that rejects 127.x.x.x and ::1.
HTML comments in tool descriptions may contain hidden instructions intended to influence LLM reasoning.
markdown_report = re.sub(r"<!--.*?-->", "", markdown_report, flags=re.DOTALL)
Remove HTML comments from description strings. Use source code comments instead.
API endpoints without rate limiting are vulnerable to brute force and denial of service.
@app.get("/{path:path}")Add rate limiting middleware to all public API endpoints.
OAuth-protected endpoints that don't validate scopes may allow unauthorized actions.
api_key: The SambaNova API key (Bearer token)
Validate OAuth scopes on every endpoint. Check that the token has required permissions.
OAuth-protected endpoints that don't validate scopes may allow unauthorized actions.
"Authorization": f"Bearer {self._api_key}",Validate OAuth scopes on every endpoint. Check that the token has required permissions.
OAuth-protected endpoints that don't validate scopes may allow unauthorized actions.
"Authorization": f"Bearer {self._api_key}",Validate OAuth scopes on every endpoint. Check that the token has required permissions.
API endpoints without rate limiting are vulnerable to brute force and denial of service.
@router.post("/message", response_model=SteeringResponse)Add rate limiting middleware to all public API endpoints.
API endpoints without rate limiting are vulnerable to brute force and denial of service.
@router.get("/plan/{session_id}", response_model=SessionPlan)Add rate limiting middleware to all public API endpoints.
API endpoints without rate limiting are vulnerable to brute force and denial of service.
@router.get("/status/{session_id}", response_model=PlanStatus)Add rate limiting middleware to all public API endpoints.
API endpoints without rate limiting are vulnerable to brute force and denial of service.
@router.get("/interactive/session/{session_id}")Add rate limiting middleware to all public API endpoints.
API endpoints without rate limiting are vulnerable to brute force and denial of service.
@router.get("/sessions")Add rate limiting middleware to all public API endpoints.
API endpoints without rate limiting are vulnerable to brute force and denial of service.
@router.get("/examples")Add rate limiting middleware to all public API endpoints.
POST/PUT/DELETE endpoints without CSRF tokens are vulnerable to cross-site request forgery.
@router.post("/message", response_model=SteeringResponse)Implement CSRF protection using tokens or SameSite cookies.
Using path.join or path.resolve with variables from user input without validation can lead to directory traversal.
file_path = os.path.join(args.input_dir, file)
Sanitize user input before passing to path.join/resolve. Use path.normalize() and check for '..' sequences.
Using Python's open() with variable paths without validation enables path traversal.
with open(f"deep_research_bench/data/test_data/raw_data/{args.model_name}.jsonl", "w", encoding="utf-8") as f:Validate and sanitize file paths using os.path.realpath() and check against allowed directories.
Direct access to sensitive files like /etc/passwd, /etc/shadow, or SSH keys indicates potential data exfiltration.
start_cmd = "/root/.jupyter/start-up.sh"
Remove direct references to sensitive system files. Use a restricted file access layer.
Backslash-based directory traversal patterns targeting Windows file systems.
description += f"- Preview: {content[:100]}...\n"Normalize path separators and apply traversal checks for both forward and backslashes.
Backslash-based directory traversal patterns targeting Windows file systems.
print("...\n")Normalize path separators and apply traversal checks for both forward and backslashes.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = llm.invoke(messages, config=config)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = model.invoke(messages)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
result = await tool.ainvoke(func_call["args"])
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
multiply_result = await tool.ainvoke(multiply_args)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
final_response = model.invoke(messages)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
final_response = model.invoke(messages)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
result = await tool.ainvoke(args)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
final_response = model.invoke(messages)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
final_response = model.invoke(messages)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
retry_response = model.invoke(messages)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
result = await tool.ainvoke(func_call["args"])
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
final_response = model.invoke(messages)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
result = await tool.ainvoke(args)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
final_response = model.invoke(messages)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
agent_response = await agent_executor.ainvoke({"input": query})Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
agent_response = await agent_executor.ainvoke({"input": query})Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = llm.invoke([message])
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = llm.invoke([message])
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = llm.invoke([message])
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = await llm_with_tool.ainvoke(messages) # Use await and ainvoke
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = await llm.ainvoke(messages)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
search_result = await tool_executor.execute_tool(
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = llm.invoke([HumanMessage(content=analysis_prompt)])
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = llm.invoke(
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = llm.invoke(
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
result = llm.invoke(
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = llm.invoke(prompt)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = llm.invoke(prompt)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = llm.invoke(prompt)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = llm.invoke(prompt)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = llm.invoke(prompt, **invoke_kwargs)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = llm.invoke(prompt) # Fallback to calling without it
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected runtime reassignment of a tool's description property to a non-literal value. Dynamically modifying tool descriptions can allow an attacker to inject misleading instructions that alter LLM behavior.
task.description = action.get("description", task.description)Use only static, hardcoded string literals for tool descriptions. Never assign descriptions from variables, user input, or external data sources.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = await llm.ainvoke(messages)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = await llm.ainvoke(messages)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = await llm.ainvoke(messages)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = await llm.ainvoke(messages)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
result = await executor.execute_tool(op_name, params)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
result = await executor.execute_tool(
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
nav_result = await executor.execute_tool(
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
pdf_result = await executor.execute_tool(
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
screenshot_result = await executor.execute_tool(
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
# snapshot_result = await executor.execute_tool(
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
# type_result = await executor.execute_tool(
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
result = await executor.execute_tool(
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
result = await executor.execute_tool(
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
result = await executor.execute_tool(
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
result = await executor.execute_tool(
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
result = await client.execute_tool(op_name, params)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
result = await tool.lc_tool.ainvoke(params, config=config)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected runtime reassignment of a tool's description property to a non-literal value. Dynamically modifying tool descriptions can allow an attacker to inject misleading instructions that alter LLM behavior.
mock_tool.description = "Echo back the message."
Use only static, hardcoded string literals for tool descriptions. Never assign descriptions from variables, user input, or external data sources.
Detected runtime reassignment of a tool's description property to a non-literal value. Dynamically modifying tool descriptions can allow an attacker to inject misleading instructions that alter LLM behavior.
assert tools[0].description == "Echo back the message."
Use only static, hardcoded string literals for tool descriptions. Never assign descriptions from variables, user input, or external data sources.
Detected runtime reassignment of a tool's description property to a non-literal value. Dynamically modifying tool descriptions can allow an attacker to inject misleading instructions that alter LLM behavior.
self.description = "Convert natural language queries to SQL and execute against uploaded databases"
Use only static, hardcoded string literals for tool descriptions. Never assign descriptions from variables, user input, or external data sources.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = llm_client.invoke([HumanMessage(content=prompt)])
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = llm_client.invoke([HumanMessage(content=prompt)])
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected runtime reassignment of a tool's description property to a non-literal value. Dynamically modifying tool descriptions can allow an attacker to inject misleading instructions that alter LLM behavior.
self.description = description
Use only static, hardcoded string literals for tool descriptions. Never assign descriptions from variables, user input, or external data sources.
Detected runtime reassignment of a tool's description property to a non-literal value. Dynamically modifying tool descriptions can allow an attacker to inject misleading instructions that alter LLM behavior.
self.description = description
Use only static, hardcoded string literals for tool descriptions. Never assign descriptions from variables, user input, or external data sources.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = await llm_with_tool.ainvoke(messages)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = await llm.ainvoke(messages)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
response = await llm.ainvoke(messages)
Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
Detected tool responses that attempt to call or invoke other tools (use_tool, call_tool, invoke, execute_tool). A poisoned tool response could trick the LLM into executing additional tools without user consent.
result = graph.invoke({"research_topic": research_topic}, config=config)Tool responses should never contain tool invocation patterns. Validate and sanitize all output to ensure it does not include cross-tool call instructions.
String literals matching known API key prefixes (sk-, ghp_, AKIA, xoxb-, etc.) or long base64-like strings may expose secrets in source code.
OPENAI_API_KEY = os.environ.get("OPENAI_API_KEY", "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")Remove hardcoded secrets from source code. Use environment variables or a secrets manager.
subprocess calls with shell=True execute commands through the shell, enabling injection attacks.
result = subprocess.run(server["install_command"], shell=True)
Set shell=False in subprocess calls and pass command as a list.
subprocess calls with shell=True execute commands through the shell, enabling injection attacks.
subprocess.run(server["run_command"], shell=True)
Set shell=False in subprocess calls and pass command as a list.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
print("\nTaking screenshot (using PDF as workaround)...")Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
screenshots_dir = os.path.join(project_root, "screenshots")
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
screenshots_dir = os.path.join(project_root, "screenshots")
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
output_file_path = os.path.join(screenshots_dir, "example_com.pdf")
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
# Copy the file to our screenshots directory
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
os.makedirs(screenshots_dir, exist_ok=True) # Ensure screenshots directory exists
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
os.makedirs(screenshots_dir, exist_ok=True) # Ensure screenshots directory exists
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
# ALSO attempt to take an actual screenshot (as a second approach)
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
print("\nAlso attempting to take a PNG screenshot...")Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
output_screenshot_path = os.path.join(screenshots_dir, "linkedin_profile.png")
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
output_screenshot_path = os.path.join(screenshots_dir, "linkedin_profile.png")
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
screenshot_result = await executor.execute_tool(
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
"mcp.playwright.browser_take_screenshot",
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
print(f"Screenshot command executed. Result: {screenshot_result}")Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
# Check if Playwright saved the screenshot somewhere and reported the location
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
if isinstance(screenshot_result, str) and "Saved as" in screenshot_result:
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
if isinstance(screenshot_result, str) and "Saved as" in screenshot_result:
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
actual_path = screenshot_result.replace("Saved as", "").strip()Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
print(f"Actual screenshot file location: {actual_path}")Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
# Copy the file to our screenshots directory
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
shutil.copy2(actual_path, output_screenshot_path)
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
if os.path.exists(output_screenshot_path):
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
print(f"Successfully copied screenshot to: {output_screenshot_path}")Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
print(f"Successfully copied screenshot to: {output_screenshot_path}")Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
print(f"Failed to copy screenshot to: {output_screenshot_path}")Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
print(f"Failed to copy screenshot to: {output_screenshot_path}")Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
print(f"Found possible screenshot at: {newest_file}")Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
shutil.copy2(newest_file, output_screenshot_path)
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
if os.path.exists(output_screenshot_path):
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
print(f"Successfully copied most recent PNG to: {output_screenshot_path}")Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
print(f"Failed to copy most recent PNG to: {output_screenshot_path}")Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
print(f"Error processing screenshot: {e}")Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
# Take a screenshot
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
print("\nTaking screenshot...")Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
screenshots_dir = os.path.join(project_root, "screenshots")
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
screenshots_dir = os.path.join(project_root, "screenshots")
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
os.makedirs(screenshots_dir, exist_ok=True)
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
"mcp.puppeteer.puppeteer_screenshot",
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
"name": "example_screenshot",
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
screenshot_path = os.path.join(screenshots_dir, "example_screenshot.png")
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
screenshot_path = os.path.join(screenshots_dir, "example_screenshot.png")
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
screenshot_path = os.path.join(screenshots_dir, "example_screenshot.png")
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
with open(screenshot_path, 'wb') as f:
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
print(f"Screenshot saved to: {screenshot_path}")Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
print(f"Error taking screenshot: {e}")Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
# Take a screenshot (optional)
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
screenshot_path = f"screenshot_{url.replace('://', '_').replace('/', '_').replace('.', '_')}.png"Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
screenshot_path = f"screenshot_{url.replace('://', '_').replace('/', '_').replace('.', '_')}.png"Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
"mcp.puppeteer.screenshot",
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
{"path": screenshot_path}Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
print(f"Screenshot saved to: {screenshot_path}")Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
"screenshot_path": screenshot_path
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Accessing clipboard contents or taking screenshots may be used to capture and exfiltrate sensitive data.
"screenshot_path": screenshot_path
Remove clipboard/screenshot access unless explicitly required by the tool's stated purpose.
Sending conversation, prompt, or session data to external storage services may leak sensitive user interactions.
self.process.stdin.write(json.dumps(message).encode() + b"\n")
Do not write session or conversation data to external storage. Keep user interaction data within the authorized session boundary.
Detected SELECT * or ORM queries without explicit field selection. Returning all columns risks exposing sensitive fields (passwords, tokens, internal IDs) to the client or LLM context.
cursor.execute("SELECT * FROM data LIMIT 3")Always specify the exact columns or fields to return. Use SELECT with explicit column names or ORM select/projection options.
Detected SELECT * or ORM queries without explicit field selection. Returning all columns risks exposing sensitive fields (passwords, tokens, internal IDs) to the client or LLM context.
cursor.execute(f"SELECT * FROM {table_name} LIMIT 3")Always specify the exact columns or fields to return. Use SELECT with explicit column names or ORM select/projection options.
Detected SELECT * or ORM queries without explicit field selection. Returning all columns risks exposing sensitive fields (passwords, tokens, internal IDs) to the client or LLM context.
sql_query = "SELECT * FROM customers WHERE state = 'CA'"
Always specify the exact columns or fields to return. Use SELECT with explicit column names or ORM select/projection options.