@byted-las/contextlake-openclaw 1.0.3 → 1.0.5

package/src/commands/cli.ts

@@ -37,29 +37,23 @@ function parseMetadata(metadata: any): Record<string, any> {
 
 export function getCliCommands(pluginConfig: ContextLakeConfig, logger: any) {
     return {
-        connectAction: async (datasource_name: string, options: any) => {
-            logger.info(`[${new Date().toISOString()}] [ContextLake] CLI connect started`, { datasource_name, options });
+        connectAction: async (datasource_name: string, url: string, options: any) => {
+            logger.info(`[${new Date().toISOString()}] [ContextLake] CLI connect started`, { datasource_name, url, options });
             try {
                 const params: ConnectParams = {
                     datasource_name,
-                    vendor: options.vendor,
+                    url,
                     endpoint: options.endpoint,
                     access_key: options.ak,
                     secret_key: options.sk,
                     region: options.region,
-                    bucket: options.bucket,
-                    prefix: options.prefix,
                     sample_rows: parseInt(options.sampleRows),
                 };
 
                 // eslint-disable-next-line no-console
                 console.log(`[contextlake connect] Connecting to datasource "${datasource_name}"...`);
                 // eslint-disable-next-line no-console
-                console.log(`  vendor: ${params.vendor}`);
-                // eslint-disable-next-line no-console
-                console.log(`  bucket: ${params.bucket}`);
-                // eslint-disable-next-line no-console
-                console.log(`  prefix: ${params.prefix}`);
+                console.log(`  url: ${params.url}`);
 
                 const result = await connectDataSource(params);
                 // eslint-disable-next-line no-console
@@ -147,13 +141,15 @@ export function getCliCommands(pluginConfig: ContextLakeConfig, logger: any) {
                 console.log('Please provide your credentials below. Press enter to keep the current value.');
                 
                 const lasApiKey = await promptForInput('LAS_API_KEY', currentCreds.LAS_API_KEY);
-                const volcengineAccessKey = await promptForInput('VOLCENGINE_ACCESS_KEY', currentCreds.VOLCENGINE_ACCESS_KEY);
-                const volcengineSecretKey = await promptForInput('VOLCENGINE_SECRET_KEY', currentCreds.VOLCENGINE_SECRET_KEY);
+                const accessKey = await promptForInput('ACCESS_KEY', currentCreds.ACCESS_KEY || currentCreds.VOLCENGINE_ACCESS_KEY);
+                const secretKey = await promptForInput('SECRET_KEY', currentCreds.SECRET_KEY || currentCreds.VOLCENGINE_SECRET_KEY);
+                const region = await promptForInput('REGION', currentCreds.REGION || currentCreds.VOLCENGINE_REGION || 'cn-beijing');
                 
                 const newCreds = {
                     LAS_API_KEY: lasApiKey,
-                    VOLCENGINE_ACCESS_KEY: volcengineAccessKey,
-                    VOLCENGINE_SECRET_KEY: volcengineSecretKey
+                    ACCESS_KEY: accessKey,
+                    SECRET_KEY: secretKey,
+                    REGION: region
                 };
                 
                 saveCredentials(newCreds);

package/src/commands/index.ts

@@ -26,6 +26,21 @@ export function registerAll(ctx: OpenClawPluginApi, logger: PluginLogger) {
         
         ctx.registerTool(tools.lasDataProfilerTool );
         logger.info(`[${new Date().toISOString()}] [ContextLake] Tool registered: ${tools.lasDataProfilerTool.name}`);
+
+        ctx.registerTool(tools.listS3ObjectsTool );
+        logger.info(`[${new Date().toISOString()}] [ContextLake] Tool registered: ${tools.listS3ObjectsTool.name}`);
+
+        ctx.registerTool(tools.readS3ObjectTool );
+        logger.info(`[${new Date().toISOString()}] [ContextLake] Tool registered: ${tools.readS3ObjectTool.name}`);
+
+        ctx.registerTool(tools.writeLanceCatalogTool );
+        logger.info(`[${new Date().toISOString()}] [ContextLake] Tool registered: ${tools.writeLanceCatalogTool.name}`);
+
+        ctx.registerTool(tools.readLanceCatalogTool );
+        logger.info(`[${new Date().toISOString()}] [ContextLake] Tool registered: ${tools.readLanceCatalogTool.name}`);
+
+        ctx.registerTool(tools.generatePresignedUrlTool );
+        logger.info(`[${new Date().toISOString()}] [ContextLake] Tool registered: ${tools.generatePresignedUrlTool.name}`);
         
         ctx.registerTool(tools.listDatasourceTool );
         logger.info(`[${new Date().toISOString()}] [ContextLake] Tool registered: ${tools.listDatasourceTool.name}`);
@@ -51,15 +66,12 @@ export function registerAll(ctx: OpenClawPluginApi, logger: PluginLogger) {
             const commands = getCliCommands(pluginConfig, logger);
             
             // connect -- data source profiling (las-data-profiler)
-            contextlake.command('connect <datasource_name>')
+            contextlake.command('connect <datasource_name> <url>')
                 .description('Connect to a data source and profile its structure, schemas, and media metadata into LanceDB')
-                .requiredOption('--vendor <vendor>', 'Data source type: volcengine | alibaba | tencent | aws | local')
                 .option('--endpoint <url>', 'S3 Endpoint URL (not needed for local)')
                 .option('--ak <credential_id>', 'Credential ID for the data source')
                 .option('--sk <credential_value>', 'Credential value for the data source')
                 .option('--region <region>', 'Region identifier (e.g. cn-beijing)')
-                .requiredOption('--bucket <bucket>', 'Bucket name (or local root directory for local vendor)')
-                .requiredOption('--prefix <prefix>', 'Path prefix to limit scan scope')
                 .option('--sample-rows <number>', 'Number of rows to sample per structured file', '100')
                 .action(commands.connectAction);
 

package/src/commands/slashcmd.ts

@@ -108,21 +108,15 @@ export function getSlashCommands(pluginConfig: ContextLakeConfig, logger: any) {
 
             logger.info(`[${new Date().toISOString()}] [ContextLake] Slash command profiler started`, { args });
             try {
-                if (args.length < 4) {
-                    return { text: `**Error:** Missing arguments. Usage: /contextlake-profiler <datasource_name> <vendor> <bucket> <prefix> [endpoint] [ak] [sk] [region]` };
+                if (args.length < 2) {
+                    return { text: `**Error:** Missing arguments. Usage: /contextlake-profiler <datasource_name> <url> [endpoint] [ak] [sk] [region]` };
                 }
                 
-                const [datasource_name, vendor, bucket, prefix, endpoint, access_key, secret_key, region] = args;
-                
-                if (!['volcengine', 'alibaba', 'tencent', 'aws', 'local'].includes(vendor)) {
-                     return { text: `**Error:** Invalid vendor. Must be one of: volcengine, alibaba, tencent, aws, local` };
-                }
+                const [datasource_name, url, endpoint, access_key, secret_key, region] = args;
                 
                 const params: ConnectParams = {
                     datasource_name,
-                    vendor: vendor as ConnectParams['vendor'],
-                    bucket,
-                    prefix,
+                    url,
                     endpoint,
                     access_key,
                     secret_key,

package/src/commands/tools.ts

@@ -3,6 +3,8 @@ import { retrieveAssets } from '../lib/actions/retrieve';
 import { listAssets, deleteAssets } from '../lib/actions/manage';
 import { connectDataSource, listDataSources } from '../lib/actions/profiler';
 import { getLasTools } from '../lib/actions/las-tools';
+import { listS3Objects, readS3Object, getPresignedUrl } from '../lib/actions/s3-tools';
+import { writeLanceCatalog, readLanceCatalog } from '../lib/actions/lance-tools';
 import { ContextLakeConfig } from '../utils/config';
 // @ts-ignore
 import type { AnyAgentTool } from 'openclaw/plugin-sdk';
@@ -14,6 +16,11 @@ export function getAgentTools(pluginConfig: ContextLakeConfig, logger: any): {
     deleteTool: AnyAgentTool;
     lasDataProfilerTool: AnyAgentTool;
     listDatasourceTool: AnyAgentTool;
+    listS3ObjectsTool: AnyAgentTool;
+    readS3ObjectTool: AnyAgentTool;
+    generatePresignedUrlTool: AnyAgentTool;
+    writeLanceCatalogTool: AnyAgentTool;
+    readLanceCatalogTool: AnyAgentTool;
     lasTools: AnyAgentTool[];
 } {
     const lasTools = getLasTools(pluginConfig, logger);
@@ -269,16 +276,10 @@ Example User Queries:
                 type: 'object',
                 properties: {
                     datasource_name: { type: 'string', description: 'Name of the data source' },
-                    vendor: { type: 'string', enum: ['volcengine', 'alibaba', 'tencent', 'aws', 'local'], description: 'Data source type' },
-                    endpoint: { type: 'string', description: 'S3 Endpoint URL (not needed for local)' },
-                    access_key: { type: 'string', description: 'Credential ID for the data source' },
-                    secret_key: { type: 'string', description: 'Credential value for the data source' },
-                    region: { type: 'string', description: 'Region identifier (e.g. cn-beijing)' },
-                    bucket: { type: 'string', description: 'Bucket name (or local root directory for local vendor)' },
-                    prefix: { type: 'string', description: 'Path prefix to limit scan scope' },
+                    url: { type: 'string', description: 'Data source URL (e.g. tos://bucket/prefix, oss://..., s3://..., file:///path)' },
                     sample_rows: { type: 'integer', description: 'Number of rows to sample per structured file' }
                 },
-                required: ['datasource_name', 'vendor', 'bucket', 'prefix'],
+                required: ['datasource_name', 'url'],
                 additionalProperties: false
             },
             
@@ -299,13 +300,177 @@ Example User Queries:
                 } catch (error: any) {
                     logger.error(`[${new Date().toISOString()}] [ContextLake] las-data-profiler skill failed`, { error: error.message, stack: error.stack });
                     return {
-                        content: [{ type: "text", text: String(error.message
-                    ) }],
-                        details: { error: error.message
-                     }
+                        content: [{ type: "text", text: String(error.message) }],
+                        details: { error: error.message }
                     } as any;
                 }
             }
+        },
+        listS3ObjectsTool: {
+            name: 'list-s3-objects',
+            label: 'List S3 Objects',
+            description: 'List objects in an S3-compatible bucket or local directory',
+            parameters: {
+                type: 'object',
+                properties: {
+                    url: { type: 'string', description: 'Data source URL (e.g. tos://bucket/prefix, oss://..., file:///path)' },
+                    vendor: { type: 'string', enum: ['volcengine', 'alibaba', 'tencent', 'aws', 'local'], description: 'Required if url is not provided' },
+                    bucket: { type: 'string', description: 'Required if url is not provided' },
+                    prefix: { type: 'string' },
+                    endpoint: { type: 'string' },
+                    maxKeys: { type: 'integer' },
+                    continuationToken: { type: 'string' }
+                },
+                required: [],
+                additionalProperties: false
+            },
+            async execute(toolCallId: string, params: any) {
+                let actualParams = params.params || params;
+                try {
+                    const result = await listS3Objects(actualParams, actualParams.prefix || '', actualParams.maxKeys, actualParams.continuationToken);
+                    return { content: [{ type: "text", text: JSON.stringify(result) }], details: result } as any;
+                } catch (e: any) {
+                    return { content: [{ type: "text", text: String(e.message) }], details: { error: e.message } } as any;
+                }
+            }
+        },
+        readS3ObjectTool: {
+            name: 'read-s3-object',
+            label: 'Read S3 Object',
+            description: 'Read the contents or headers of an S3 object',
+            parameters: {
+                type: 'object',
+                properties: {
+                    url: { type: 'string', description: 'Full URL to the object (e.g. tos://bucket/path/to/key.txt)' },
+                    vendor: { type: 'string', enum: ['volcengine', 'alibaba', 'tencent', 'aws', 'local'], description: 'Required if url is not provided' },
+                    bucket: { type: 'string', description: 'Required if url is not provided' },
+                    key: { type: 'string', description: 'Required if url is not provided' },
+                    endpoint: { type: 'string' },
+                    maxBytes: { type: 'integer' }
+                },
+                required: [],
+                additionalProperties: false
+            },
+            async execute(toolCallId: string, params: any) {
+                let actualParams = params.params || params;
+                try {
+                    // Extract key from url if provided
+                    let key = actualParams.key;
+                    if (actualParams.url && !key) {
+                        try {
+                            if (actualParams.url.startsWith('file://')) {
+                                // Key is not strictly needed for file://, bucket contains the path in parseS3Url
+                                key = ''; 
+                            } else {
+                                const parsedUrl = new URL(actualParams.url);
+                                key = parsedUrl.pathname.replace(/^\//, '');
+                            }
+                        } catch (e) {
+                            // let it fail in readS3Object
+                        }
+                    }
+                    if (!key && !actualParams.url?.startsWith('file://')) {
+                        throw new Error('key is required or must be part of the url');
+                    }
+                    const buf = await readS3Object(actualParams, key, actualParams.maxBytes);
+                    // Return as base64 string
+                    return { content: [{ type: "text", text: buf.toString('base64') }], details: { length: buf.length } } as any;
+                } catch (e: any) {
+                    return { content: [{ type: "text", text: String(e.message) }], details: { error: e.message } } as any;
+                }
+            }
+        },
+        generatePresignedUrlTool: {
+            name: 'generate-presigned-url',
+            label: 'Generate Presigned URL',
+            description: 'Generate a presigned HTTP URL for an S3/TOS object, allowing temporary public access',
+            parameters: {
+                type: 'object',
+                properties: {
+                    url: { type: 'string', description: 'Full URL to the object (e.g. tos://bucket/path/to/key.txt)' },
+                    vendor: { type: 'string', enum: ['volcengine', 'alibaba', 'tencent', 'aws', 'local'], description: 'Required if url is not provided' },
+                    bucket: { type: 'string', description: 'Required if url is not provided' },
+                    key: { type: 'string', description: 'Required if url is not provided' },
+                    endpoint: { type: 'string' },
+                    expiresIn: { type: 'integer', description: 'Expiration time in seconds (default 3600)' }
+                },
+                required: [],
+                additionalProperties: false
+            },
+            async execute(toolCallId: string, params: any) {
+                let actualParams = params.params || params;
+                try {
+                    let key = actualParams.key;
+                    if (actualParams.url && !key) {
+                        try {
+                            if (actualParams.url.startsWith('file://')) {
+                                key = ''; 
+                            } else {
+                                const parsedUrl = new URL(actualParams.url);
+                                key = parsedUrl.pathname.replace(/^\//, '');
+                            }
+                        } catch (e) {
+                            // let it fail in getPresignedUrl
+                        }
+                    }
+                    if (!key && !actualParams.url?.startsWith('file://')) {
+                        throw new Error('key is required or must be part of the url');
+                    }
+                    const url = await getPresignedUrl(actualParams, key, actualParams.expiresIn);
+                    return { content: [{ type: "text", text: url }], details: { url } } as any;
+                } catch (e: any) {
+                    return { content: [{ type: "text", text: String(e.message) }], details: { error: e.message } } as any;
+                }
+            }
+        },
+        writeLanceCatalogTool: {
+            name: 'write-lance-catalog',
+            label: 'Write LanceDB Catalog',
+            description: 'Write an array of file records into a local LanceDB table',
+            parameters: {
+                type: 'object',
+                properties: {
+                    db_path: { type: 'string' },
+                    table_name: { type: 'string' },
+                    records: { type: 'array', items: { type: 'object' } }
+                },
+                required: ['db_path', 'table_name', 'records'],
+                additionalProperties: false
+            },
+            async execute(toolCallId: string, params: any) {
+                let actualParams = params.params || params;
+                try {
+                    await writeLanceCatalog(actualParams);
+                    return { content: [{ type: "text", text: "Successfully wrote records to LanceDB" }], details: { count: actualParams.records.length } } as any;
+                } catch (e: any) {
+                    return { content: [{ type: "text", text: String(e.message) }], details: { error: e.message } } as any;
+                }
+            }
+        },
+        readLanceCatalogTool: {
+            name: 'read-lance-catalog',
+            label: 'Read LanceDB Catalog',
+            description: 'Read records from a local LanceDB table for validation or ingestion processes',
+            parameters: {
+                type: 'object',
+                properties: {
+                    db_path: { type: 'string', description: 'Path to the local LanceDB database' },
+                    table_name: { type: 'string', description: 'Name of the table to read' },
+                    limit: { type: 'integer', description: 'Maximum number of records to return' },
+                    filter: { type: 'string', description: 'SQL-like filter string (e.g., "category = \'structured\'")' }
+                },
+                required: ['db_path', 'table_name'],
+                additionalProperties: false
+            },
+            async execute(toolCallId: string, params: any) {
+                let actualParams = params.params || params;
+                try {
+                    const results = await readLanceCatalog(actualParams);
+                    return { content: [{ type: "text", text: JSON.stringify(results) }], details: { count: results.length, data: results } } as any;
+                } catch (e: any) {
+                    return { content: [{ type: "text", text: String(e.message) }], details: { error: e.message } } as any;
+                }
+            }
         }
     };
 }

package/src/lib/actions/lance-tools.ts

@@ -0,0 +1,58 @@
+import * as lancedb from '@lancedb/lancedb';
+import * as fs from 'fs';
+
+export interface LanceWriteParams {
+    db_path: string;
+    table_name: string;
+    records: any[];
+}
+
+export interface LanceReadParams {
+    db_path: string;
+    table_name: string;
+    limit?: number;
+    filter?: string;
+}
+
+export async function writeLanceCatalog(params: LanceWriteParams) {
+    if (!params.records || params.records.length === 0) {
+        return;
+    }
+
+    const db = await lancedb.connect(params.db_path);
+    const tableNames = await db.tableNames();
+    
+    if (tableNames.includes(params.table_name)) {
+        const table = await db.openTable(params.table_name);
+        await table.add(params.records);
+    } else {
+        await db.createTable(params.table_name, params.records);
+    }
+}
+
+export async function readLanceCatalog(params: LanceReadParams) {
+    if (!fs.existsSync(params.db_path)) {
+        throw new Error(`Database not found at ${params.db_path}`);
+    }
+
+    const db = await lancedb.connect(params.db_path);
+    const tableNames = await db.tableNames();
+    
+    if (!tableNames.includes(params.table_name)) {
+        throw new Error(`Table '${params.table_name}' not found in database`);
+    }
+
+    const table = await db.openTable(params.table_name);
+    let query = table.query();
+    
+    if (params.filter) {
+        query = query.where(params.filter);
+    }
+    
+    if (params.limit && params.limit > 0) {
+        query = query.limit(params.limit);
+    }
+    
+    const results = await query.toArray();
+    return results;
+}

package/src/lib/actions/las-tools.ts

@@ -1,11 +1,36 @@
 import { LasApiClient } from './las-api';
 import { ContextLakeConfig } from '../../utils/config';
+import { getPresignedUrl } from './s3-tools';
 // @ts-ignore
 import type { AnyAgentTool } from 'openclaw/plugin-sdk';
 
 export function getLasTools(pluginConfig: ContextLakeConfig, logger: any): AnyAgentTool[] {
     const apiClient = new LasApiClient(pluginConfig, logger);
 
+    const processUrl = async (url: string): Promise<string> => {
+        if (!url) return url;
+        if (url.startsWith('tos://')) {
+            // LAS operators prefer native tos:// paths when supported, leave as is
+            return url;
+        } else if (url.startsWith('oss://') || url.startsWith('s3://') || url.startsWith('cos://') || url.startsWith('file://')) {
+            // Need presigned HTTP url for other vendors
+            logger.info(`[LasTools] Presigning URL for vendor: ${url}`);
+            try {
+                // If it's a file:// we also presign it to file:// which might not be supported by remote LAS, 
+                // but local files typically need to be uploaded to TOS first. We'll leave file:// to fail or be handled elsewhere.
+                if (url.startsWith('file://')) return url;
+                
+                const urlParts = new URL(url);
+                const key = urlParts.pathname.replace(/^\//, '');
+                return await getPresignedUrl({ url }, key, 3600);
+            } catch (e: any) {
+                logger.warn(`[LasTools] Failed to presign URL: ${url}`, { error: e.message });
+                return url; // fallback to original
+            }
+        }
+        return url;
+    };
+
     const callApi = async (method: string, args: any[]) => {
         try {
             // @ts-ignore
@@ -35,6 +60,9 @@ Parameters in data:
                 required: ['data']
             },
             async execute(toolCallId: string, params: any) {
+                if (params.data?.image) {
+                    params.data.image = await processUrl(params.data.image);
+                }
                 return await callApi('process', ['las_image_resample', params.data]);
             }
         },
@@ -55,6 +83,9 @@ Parameters in data:
                 required: ['data']
             },
             async execute(toolCallId: string, params: any) {
+                if (params.data?.input_path) {
+                    params.data.input_path = await processUrl(params.data.input_path);
+                }
                 return await callApi('process', ['las_audio_extract_and_split', params.data]);
             }
         },
@@ -73,6 +104,9 @@ Parameters in data:
                 required: ['data']
             },
             async execute(toolCallId: string, params: any) {
+                if (params.data?.input_path) {
+                    params.data.input_path = await processUrl(params.data.input_path);
+                }
                 return await callApi('process', ['las_audio_convert', params.data]);
             }
         },
@@ -91,6 +125,9 @@ Parameters in data:
                 required: ['data']
             },
             async execute(toolCallId: string, params: any) {
+                if (params.data?.audio?.url) {
+                    params.data.audio.url = await processUrl(params.data.audio.url);
+                }
                 return await callApi('submitAndPoll', ['las_asr_pro', params.data]);
             }
         },
@@ -107,6 +144,9 @@ Parameters in data:
                 required: ['data']
             },
             async execute(toolCallId: string, params: any) {
+                if (params.data?.audio?.url) {
+                    params.data.audio.url = await processUrl(params.data.audio.url);
+                }
                 return await callApi('submitAndPoll', ['las_seed_2_0', params.data]);
             }
         },
@@ -130,6 +170,13 @@ Parameters:
                 required: ['model', 'input']
             },
             async execute(toolCallId: string, params: any) {
+                if (params.input && Array.isArray(params.input)) {
+                    for (const item of params.input) {
+                        if (item.type === 'image_url' && item.image_url?.url) {
+                            item.image_url.url = await processUrl(item.image_url.url);
+                        }
+                    }
+                }
                 return await callApi('multimodalEmbedding', [
                     params.model,
                     params.input,
@@ -153,6 +200,9 @@ Parameters in data:
                 required: ['data']
             },
             async execute(toolCallId: string, params: any) {
+                if (params.data?.video_url) {
+                    params.data.video_url = await processUrl(params.data.video_url);
+                }
                 return await callApi('submitAndPoll', ['las_long_video_understand', params.data]);
             }
         },
@@ -171,6 +221,9 @@ Parameters in data:
                 required: ['data']
             },
             async execute(toolCallId: string, params: any) {
+                if (params.data?.url) {
+                    params.data.url = await processUrl(params.data.url);
+                }
                 return await callApi('submitAndPoll', ['las_pdf_parse_doubao', params.data]);
             }
         },
@@ -189,6 +242,9 @@ Parameters in data:
                 required: ['data']
             },
             async execute(toolCallId: string, params: any) {
+                if (params.data?.video_url) {
+                    params.data.video_url = await processUrl(params.data.video_url);
+                }
                 return await callApi('submitAndPoll', ['las_video_resize', params.data]);
             }
         }