# Send powerdrill-data-analysis to your agent Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually. ## Fast path - Download the package from Yavira. - Extract it into a folder your agent can access. - Paste one of the prompts below and point your agent at the extracted folder. ## Suggested prompts ### New install ```text I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Then review README.md for any prerequisites, environment setup, or post-install checks. Tell me what you changed and call out any manual steps you could not complete. ``` ### Upgrade existing ```text I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run. ``` ## Machine-readable fields ```json { "schemaVersion": "1.0", "item": { "slug": "powerdrill-data-analysis-skill", "name": "powerdrill-data-analysis", "source": "tencent", "type": "skill", "category": "数据分析", "sourceUrl": "https://clawhub.ai/javainthinking/powerdrill-data-analysis-skill", "canonicalUrl": "https://clawhub.ai/javainthinking/powerdrill-data-analysis-skill", "targetPlatform": "OpenClaw" }, "install": { "downloadUrl": "/downloads/powerdrill-data-analysis-skill", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=powerdrill-data-analysis-skill", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "packageFormat": "ZIP package", "primaryDoc": "SKILL.md", "includedAssets": [ "README.md", "SKILL.md", "scripts/powerdrill_client.py" ], "downloadMode": "redirect", "sourceHealth": { "source": "tencent", "slug": "powerdrill-data-analysis-skill", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-04-30T15:33:36.412Z", "expiresAt": "2026-05-07T15:33:36.412Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=powerdrill-data-analysis-skill", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=powerdrill-data-analysis-skill", "contentDisposition": "attachment; filename=\"powerdrill-data-analysis-skill-1.0.0.zip\"", "redirectLocation": null, "bodySnippet": null, "slug": "powerdrill-data-analysis-skill" }, "scope": "item", "summary": "Item download looks usable.", "detail": "Yavira can redirect you to the upstream package for this item.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/powerdrill-data-analysis-skill" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] } }, "links": { "detailUrl": "https://openagent3.xyz/skills/powerdrill-data-analysis-skill", "downloadUrl": "https://openagent3.xyz/downloads/powerdrill-data-analysis-skill", "agentUrl": "https://openagent3.xyz/skills/powerdrill-data-analysis-skill/agent", "manifestUrl": "https://openagent3.xyz/skills/powerdrill-data-analysis-skill/agent.json", "briefUrl": "https://openagent3.xyz/skills/powerdrill-data-analysis-skill/agent.md" } } ``` ## Documentation ### Powerdrill Data Analysis Skill Analyze data using the Powerdrill API via the Python client at scripts/powerdrill_client.py. All operations use the Powerdrill REST API v2 (https://ai.data.cloud/api). ### Prerequisites & Setup Before using any Powerdrill functions, the user must have: A Powerdrill Teamspace - Created by following: https://www.youtube.com/watch?v=I-0yGD9HeDw API Credentials - Obtained by following: https://www.youtube.com/watch?v=qs-GsUgjb1g Set these environment variables before running any script: export POWERDRILL_USER_ID="your_user_id" export POWERDRILL_PROJECT_API_KEY="your_project_api_key" The only Python dependency is requests. Install with: pip install requests If a call fails with an authentication error, verify the two environment variables are set and the API key is valid. ### How to Use Import the client module and call functions directly. All functions read credentials from the environment automatically. import sys sys.path.insert(0, "/absolute/path/to/scripts") # adjust to actual location from powerdrill_client import * Or run via CLI: python scripts/powerdrill_client.py [args] ### Datasets list_datasets(page_number=1, page_size=10, search=None) -> dict List datasets in the user's account. Typically the first step in any workflow. result = list_datasets(search="sales") for ds in result["data"]["records"]: print(ds["id"], ds["name"]) create_dataset(name, description="") -> dict Create a new empty dataset. Returns {"data": {"id": "dset-..."}}. ds = create_dataset("Q4 Sales Data", "Quarterly sales analysis") dataset_id = ds["data"]["id"] get_dataset_overview(dataset_id) -> dict Get dataset summary, exploration questions, and keywords. Use after data sources are synced. overview = get_dataset_overview(dataset_id) print(overview["data"]["summary"]) for q in overview["data"]["exploration_questions"]: print(f" - {q}") get_dataset_status(dataset_id) -> dict Check how many data sources are synced/syncing/invalid. status = get_dataset_status(dataset_id) # status["data"] = {"synched_count": 3, "synching_count": 0, "invalid_count": 0} delete_dataset(dataset_id) -> dict Permanently delete a dataset and all its data sources. Irreversible - always confirm with the user first. ### Data Sources list_data_sources(dataset_id, page_number=1, page_size=10, status=None) -> dict List files within a dataset. Filter by status: synched, synching, invalid. sources = list_data_sources(dataset_id, status="synched") create_data_source(dataset_id, name, *, url=None, file_object_key=None) -> dict Create a data source from a public URL or an uploaded file key. Provide exactly one of url or file_object_key. # From public URL ds = create_data_source(dataset_id, "report.pdf", url="https://example.com/report.pdf") # From uploaded file (see upload_local_file) ds = create_data_source(dataset_id, "data.csv", file_object_key=key) upload_local_file(file_path) -> str Upload a local file via multipart upload. Returns file_object_key for use with create_data_source(). Supported formats: .csv, .tsv, .md, .mdx, .json, .txt, .pdf, .pptx, .docx, .xls, .xlsx upload_and_create_data_source(dataset_id, file_path) -> dict Convenience function: uploads a local file then creates the data source in one call. result = upload_and_create_data_source(dataset_id, "/path/to/sales.csv") datasource_id = result["data"]["id"] wait_for_dataset_sync(dataset_id, max_attempts=30, delay_seconds=3.0) -> dict Poll until all data sources in the dataset are synced. Raises RuntimeError on timeout or if invalid sources are detected. upload_and_create_data_source(dataset_id, "data.csv") wait_for_dataset_sync(dataset_id) # blocks until synced ### Sessions create_session(name, output_language="AUTO", job_mode="AUTO", max_contextual_job_history=10) -> dict Create an analysis session. Required before running jobs. session = create_session("Sales Analysis Session") session_id = session["data"]["id"] list_sessions(page_number=1, page_size=10, search=None) -> dict List existing sessions. Use to find a previous session for resumption. delete_session(session_id) -> dict Delete a session. Use during cleanup after analysis is complete. ### Jobs (Data Analysis) create_job(session_id, question, dataset_id=None, datasource_ids=None, stream=False, output_language="AUTO", job_mode="AUTO") -> dict Run a natural-language analysis query. This is the core analysis function. Non-streaming (default): returns full response with all blocks. result = create_job(session_id, "What are the top 5 products by revenue?", dataset_id=dataset_id) for block in result["data"]["blocks"]: if block["type"] == "MESSAGE": print(block["content"]) elif block["type"] == "TABLE": print(f"Table: {block['content']['url']}") elif block["type"] == "IMAGE": print(f"Chart: {block['content']['url']}") Streaming: returns parsed result with accumulated text and separate blocks. result = create_job(session_id, "Summarize trends", dataset_id=dataset_id, stream=True) print(result["text"]) # accumulated MESSAGE text for b in result["blocks"]: # TABLE, IMAGE, etc. print(b["type"], b["content"]) Response block types: MESSAGE - Analytical text CODE - Code snippets (Markdown) TABLE - {name, url, expires_at} - download before expiration IMAGE - {name, url, expires_at} - download before expiration SOURCES - Citation references QUESTIONS - Suggested follow-up questions CHART_INFO - Chart configuration and data ### Cleanup cleanup(session_id=None, dataset_id=None) -> None Delete session and/or dataset after analysis. Always call this when done. cleanup(session_id=session_id, dataset_id=dataset_id) cleanup_session(session_id) -> None / cleanup_dataset(dataset_id) -> None Delete individual resources. Errors are logged but not raised. ### Full analysis workflow (upload, analyze, cleanup) from powerdrill_client import * # 1. Create dataset and upload data ds = create_dataset("My Analysis") dataset_id = ds["data"]["id"] upload_and_create_data_source(dataset_id, "/path/to/data.csv") wait_for_dataset_sync(dataset_id) # 2. Create session and run analysis session = create_session("Analysis Session") session_id = session["data"]["id"] result = create_job(session_id, "What are the key trends?", dataset_id=dataset_id) for block in result["data"]["blocks"]: if block["type"] == "MESSAGE": print(block["content"]) # 3. Ask follow-up questions (same session for context) result = create_job(session_id, "Break this down by region", dataset_id=dataset_id) # 4. Cleanup when done cleanup(session_id=session_id, dataset_id=dataset_id) ### Analyze existing dataset from powerdrill_client import * # 1. Find the dataset datasets = list_datasets(search="sales") dataset_id = datasets["data"]["records"][0]["id"] # 2. Explore it overview = get_dataset_overview(dataset_id) print(overview["data"]["summary"]) # 3. Create session and analyze session = create_session("Quick Analysis") session_id = session["data"]["id"] result = create_job(session_id, overview["data"]["exploration_questions"][0], dataset_id=dataset_id) # 4. Cleanup session when done (keep dataset) cleanup_session(session_id) ### CLI usage # List datasets python scripts/powerdrill_client.py list-datasets --search "sales" # Create dataset + upload file python scripts/powerdrill_client.py create-dataset "Test Data" python scripts/powerdrill_client.py upload-file dset-xxx /path/to/file.csv python scripts/powerdrill_client.py wait-sync dset-xxx # Create session and run a job python scripts/powerdrill_client.py create-session "My Session" python scripts/powerdrill_client.py create-job SESSION_ID "Summarize the data" --dataset-id dset-xxx # Cleanup python scripts/powerdrill_client.py cleanup --session-id SESSION_ID --dataset-id dset-xxx ### Error Handling Authentication errors: Verify POWERDRILL_USER_ID and POWERDRILL_PROJECT_API_KEY. Direct the user to the setup videos above. Dataset not found: Re-run list_datasets() to verify the ID. The dataset may have been deleted. Job execution failure: Ensure the dataset has at least one synced data source (wait_for_dataset_sync()). Retry with a rephrased question. Upload timeout: wait_for_dataset_sync() polls up to 30 attempts (90s). Use get_dataset_status() to check manually. Invalid data sources: Check file format is supported. Re-upload with correct file type. Rate limiting: Wait before retrying. Space out rapid sequential API calls. ### Important Notes Always create a session before running analysis jobs Always call cleanup() to delete sessions and datasets after analysis is complete Sessions maintain conversational context - reuse the same session for related follow-up questions TABLE and IMAGE URLs in job responses expire - download or present results promptly Call wait_for_dataset_sync() after uploading files, before running analysis Dataset and session names are limited to 128 characters Supported file formats: .csv, .tsv, .md, .mdx, .json, .txt, .pdf, .pptx, .docx, .xls, .xlsx ## Trust - Source: tencent - Verification: Indexed source record - Publisher: javainthinking - Version: 1.0.0 ## Source health - Status: healthy - Item download looks usable. - Yavira can redirect you to the upstream package for this item. - Health scope: item - Reason: direct_download_ok - Checked at: 2026-04-30T15:33:36.412Z - Expires at: 2026-05-07T15:33:36.412Z - Recommended action: Download for OpenClaw ## Links - [Detail page](https://openagent3.xyz/skills/powerdrill-data-analysis-skill) - [Send to Agent page](https://openagent3.xyz/skills/powerdrill-data-analysis-skill/agent) - [JSON manifest](https://openagent3.xyz/skills/powerdrill-data-analysis-skill/agent.json) - [Markdown brief](https://openagent3.xyz/skills/powerdrill-data-analysis-skill/agent.md) - [Download page](https://openagent3.xyz/downloads/powerdrill-data-analysis-skill)