{"id":"5dcc7304-9749-4d20-9eb1-3344541a3487","entityType":"agent","slug":"clawhub-cellcog-data-analysis-cellcog","name":"Data Analysis","canonicalUrl":"https://www.xpersona.co/agent/clawhub-cellcog-data-analysis-cellcog","canonicalPath":"/agent/clawhub-cellcog-data-analysis-cellcog","generatedAt":"2026-10-09T12:52:29.092Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-10-09T04:25:17.067Z","emptyReason":null},"description":"AI data analysis and visualization powered by CellCog. Data cleaning, exploratory analysis, hypothesis testing, statistical reports, ML model evaluation, dataset profiling, charts, dashboards. Upload CSVs and get analysis back with full Python access. Skill: Data Analysis Owner: cellcog Summary: AI data analysis and visualization powered by CellCog. Data cleaning, exploratory analysis, hypothesis testing, statistical reports, ML model evaluation, dataset profiling, charts, dashboards. Upload CSVs and get analysis back with full Python access. Tags: latest:1.0.16 Version history: v1.0.16 | 2026-08-24T02:01:10.439Z | user Content updated. v1.0.15 | 2026-08-24T01:45:","descriptionLabel":"Technical summary","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 5.1K downloads reported by the source. Last updated 10/9/2026.","installCommand":"clawhub skill install s176q1btpn094ats9b4f9hgfwx83knpk:data-analysis-cellcog","sourceUrl":"https://clawhub.ai/cellcog/data-analysis-cellcog","homepage":"https://clawhub.ai/cellcog/skills/data-analysis-cellcog","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/cellcog/data-analysis-cellcog","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/cellcog/skills/data-analysis-cellcog","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":74,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"AI data analysis and visualization powered by CellCog. Data cleaning, exploratory analysis, hypothesis testing, statistical reports, ML model evaluation, datase"},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-09T04:25:17.067Z","emptyReason":null},"protocols":[{"protocol":"OPENCLEW","label":"OpenClaw","status":"self-declared","notes":"Declared in the public agent profile."}],"capabilities":[],"verifiedCount":0,"selfDeclaredCount":1,"capabilityMatrix":{"rows":[{"key":"OPENCLEW","type":"protocol","support":"unknown","confidenceSource":"profile","notes":"Listed on profile"}],"flattenedTokens":"protocol:OPENCLEW|unknown|profile"}},"adoption":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-09T04:25:17.067Z","emptyReason":null},"stars":null,"forks":null,"downloads":5066,"packageName":null,"latestVersion":"1.0.16","tractionLabel":"5.1K downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-09T04:25:17.067Z","emptyReason":null},"lastUpdatedAt":"2026-10-09T04:25:17.067Z","lastCrawledAt":"2026-10-09T04:25:17.067Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-10T04:25:17.067Z","lastVerifiedAt":null,"highlights":[{"version":"1.0.16","createdAt":"2026-08-24T02:01:10.439Z","changelog":"Content updated.","fileCount":3,"zipByteSize":5982},{"version":"1.0.15","createdAt":"2026-08-24T01:45:28.588Z","changelog":"Content updated.","fileCount":3,"zipByteSize":5945},{"version":"1.0.14","createdAt":"2026-08-03T06:05:52.041Z","changelog":"Content updated.","fileCount":3,"zipByteSize":5942},{"version":"1.0.13","createdAt":"2026-08-02T20:46:35.540Z","changelog":"Display title updated.","fileCount":3,"zipByteSize":5854},{"version":"1.0.12","createdAt":"2026-07-17T03:40:45.480Z","changelog":"- Skill name updated from \"data-cog\" to \"data-analysis-cellcog\" for consistency. - Description and section headers revised for clarity and accuracy. - Minor documentation cleanups in SKILL.md; improved formatting and language. - Removed outdated file: skill-card.md.","fileCount":3,"zipByteSize":5988},{"version":"1.0.11","createdAt":"2026-04-23T06:18:48.710Z","changelog":"- Added `requires` field to metadata, specifying dependencies on Python 3 and the CELLCOG_API_KEY environment variable for improved skill setup clarity. - No changes to features or code behavior; documentation updated to reflect requirements.","fileCount":3,"zipByteSize":6010},{"version":"1.0.10","createdAt":"2026-04-14T17:32:07.438Z","changelog":"- Simplified and shortened the SKILL.md description for clarity. - Updated agent usage instructions for greater accuracy, clarifying \"OpenClaw\" vs. other agents. - No functional or code changes; documentation only.","fileCount":2,"zipByteSize":4751},{"version":"1.0.9","createdAt":"2026-04-13T01:00:40.063Z","changelog":"- Updated skill description for improved clarity and added latest benchmark achievement (#1 on DeepResearch Bench, Apr 2026). - Enhanced usage section with explicit CellCogClient example for agent provider configuration. - Minor edits for conciseness and modernized language throughout documentation. - No code or interface changes; documentation only.","fileCount":2,"zipByteSize":4793}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s176q1btpn094ats9b4f9hgfwx83knpk:data-analysis-cellcog","setupComplexity":"low","setupSteps":["Setup complexity is classified as HIGH. You must provision dedicated cloud infrastructure or an isolated VM. Do not run this directly on your local workstation.","Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data."],"contract":{"contractStatus":"missing","authModes":[],"requires":[],"forbidden":[],"supportsMcp":false,"supportsA2a":false,"supportsStreaming":false,"inputSchemaRef":null,"outputSchemaRef":null,"dataRegion":null,"contractUpdatedAt":null,"sourceUpdatedAt":null,"freshnessSeconds":null},"invocationGuide":{"preferredApi":{"snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-cellcog-data-analysis-cellcog/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-cellcog-data-analysis-cellcog/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-cellcog-data-analysis-cellcog/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-cellcog-data-analysis-cellcog/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-cellcog-data-analysis-cellcog/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-cellcog-data-analysis-cellcog/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-09T12:52:29.087Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-cellcog-data-analysis-cellcog/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-cellcog-data-analysis-cellcog/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-cellcog-data-analysis-cellcog/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-cellcog-data-analysis-cellcog/trust"}},"reliability":{"evidence":{"source":"runtime-metrics","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No trust, reliability, or runtime telemetry is available."},"trust":{"status":"unavailable","handshakeStatus":"UNKNOWN","verificationFreshnessHours":null,"reputationScore":null,"p95LatencyMs":null,"successRate30d":null,"fallbackRate":null,"attempts30d":null,"trustUpdatedAt":null,"trustConfidence":"unknown","sourceUpdatedAt":null,"freshnessSeconds":null},"decisionGuardrails":{"doNotUseIf":["Contract metadata is missing or unavailable for deterministic execution."],"safeUseWhen":[],"riskFlags":["missing_or_unavailable_contract","trust_data_unavailable","schema_references_missing"],"operationalConfidence":"low"},"executionMetrics":{"observedLatencyMsP50":null,"observedLatencyMsP95":null,"estimatedCostUsd":null,"uptime30d":null,"rateLimitRpm":null,"rateLimitBurst":null,"lastVerifiedAt":null,"verificationSource":null},"runtimeMetrics":{"successRate":null,"avgLatencyMs":null,"avgCostUsd":null,"hallucinationRate":null,"retryRate":null,"disputeRate":null,"p50Latency":null,"p95Latency":null,"lastUpdated":null}},"benchmarks":{"evidence":{"source":"no-benchmark-data","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No benchmark suites or observed failure patterns are available."},"suites":[],"failurePatterns":[]},"artifacts":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"high","updatedAt":"2026-10-09T04:25:17.067Z","emptyReason":null},"readme":"Skill: Data Analysis\n\nOwner: cellcog\n\nSummary: AI data analysis and visualization powered by CellCog. Data cleaning, exploratory analysis, hypothesis testing, statistical reports, ML model evaluation, dataset profiling, charts, dashboards. Upload CSVs and get analysis back with full Python access.\n\nTags: latest:1.0.16\n\nVersion history:\n\nv1.0.16 | 2026-08-24T02:01:10.439Z | user\n\nContent updated.\n\nv1.0.15 | 2026-08-24T01:45:28.588Z | user\n\nContent updated.\n\nv1.0.14 | 2026-08-03T06:05:52.041Z | user\n\nContent updated.\n\nv1.0.13 | 2026-08-02T20:46:35.540Z | user\n\nDisplay title updated.\n\nv1.0.12 | 2026-07-17T03:40:45.480Z | auto\n\n- Skill name updated from \"data-cog\" to \"data-analysis-cellcog\" for consistency.\n- Description and section headers revised for clarity and accuracy.\n- Minor documentation cleanups in SKILL.md; improved formatting and language.\n- Removed outdated file: skill-card.md.\n\nv1.0.11 | 2026-04-23T06:18:48.710Z | auto\n\n- Added `requires` field to metadata, specifying dependencies on Python 3 and the CELLCOG_API_KEY environment variable for improved skill setup clarity.\n- No changes to features or code behavior; documentation updated to reflect requirements.\n\nv1.0.10 | 2026-04-14T17:32:07.438Z | auto\n\n- Simplified and shortened the SKILL.md description for clarity.\n- Updated agent usage instructions for greater accuracy, clarifying \"OpenClaw\" vs. other agents.\n- No functional or code changes; documentation only.\n\nv1.0.9 | 2026-04-13T01:00:40.063Z | auto\n\n- Updated skill description for improved clarity and added latest benchmark achievement (#1 on DeepResearch Bench, Apr 2026).\n- Enhanced usage section with explicit CellCogClient example for agent provider configuration.\n- Minor edits for conciseness and modernized language throughout documentation.\n- No code or interface changes; documentation only.\n\nv1.0.8 | 2026-04-12T23:35:17.992Z | auto\n\n**Changelog for data-cog v1.0.8**\n\n- Major rewrite of SKILL.md for clarity and brevity: description and documentation made much more concise.\n- Simplified feature lists and usage instructions, focusing on core data analysis, visualization, and supported formats.\n- Updated usage examples for compatibility with new agents and platforms.\n- Removed excessive marketing, benchmark claims, and redundant text.\n- Improved section organization for easier reading and faster onboarding.\n\nv1.0.7 | 2026-04-08T05:57:51.920Z | auto\n\n- Expanded and restructured documentation in SKILL.md for improved clarity and detail.\n- Added more example prompts and use cases, showing a wide range of data analysis tasks.\n- Clarified what sets Data Cog apart from other AI tools: results, not just code.\n- Listed supported input and output formats more explicitly.\n- Updated tool description, highlighting performance on DeepResearch Bench (Apr 2026).\n\nv1.0.6 | 2026-04-06T05:07:04.178Z | auto\n\n- Major documentation update: SKILL.md rewritten for brevity and clarity.\n- Simplified description and usage instructions to focus on main features.\n- Added concise summaries of internal capabilities (Python libraries, dashboards, spreadsheet, PDF output).\n- Streamlined data analysis examples and recommended chat modes.\n- Linked to related skills for broader use cases.\n\nv1.0.5 | 2026-04-03T01:43:20.032Z | auto\n\n- Added detailed usage instructions for OpenClaw agents and other agents in the Prerequisites section.\n- Updated the sample Python code to clarify blocking vs. fire-and-forget modes.\n- Improved references to the cellcog skill for SDK details and usage guidance.\n- No changes to core features or supported data types; documentation only.\n\nv1.0.4 | 2026-04-03T00:07:43.199Z | auto\n\n- Updated SKILL.md for a more streamlined \"Quick start\" example and clarified SDK instructions.\n- Added references to the cellcog skill for full SDK API details, delivery modes, and advanced file handling.\n- Removed the previous detailed Python example in favor of a more general usage pattern.\n- No changes to core functionalityβdocumentation update only.\n\nv1.0.3 | 2026-04-02T03:33:42.053Z | auto\n\n- Updated DeepResearch Bench rating date from Feb 2026 to Apr 2026 in the description and introduction.\n- No feature or functional changes; documentation update only.\n\nv1.0.2 | 2026-03-27T04:17:16.538Z | auto\n\n- Updated skill description to clarify features and simplify language.\n- Added supported operating systems (darwin, linux, windows) in metadata.\n- Included a homepage link in the metadata.\n- No changes to functionality; documentation only.\n\nv1.0.1 | 2026-02-11T01:41:01.965Z | user\n\n- Added clear author and dependency metadata to SKILL.md.\n- Changed prerequisite wording to reference the `cellcog` skill directly.\n- No functional changes; documentation improvements only.\n\nv1.0.0 | 2026-02-08T00:43:08.712Z | user\n\n- Initial release of Data-Cog skill.\n- Enables analysis of messy CSVs and other data files with minimal prompts, returning structured insights (charts, dashboards, reports, and clean data).\n- Provides full Python access for tasks such as data cleaning, exploratory analysis, visualization, hypothesis testing, ML model evaluation, and dataset profiling.\n- Focuses on delivering actual answers and visual summaries instead of just sharing code.\n- Supports multiple data and output formats, including CSV, Excel, JSON, Parquet, and SQL exports.\n- Requires the CellCog mothership skill for SDK and API usage.\n\nArchive index:\n\nArchive v1.0.16: 3 files, 5982 bytes\n\nFiles: skill-card.md (2190b), SKILL.md (10331b), _meta.json (141b)\n\nFile v1.0.16:SKILL.md\n\n---\nname: data-analysis-cellcog\ndescription: \"AI data analysis and visualization powered by CellCog. Data cleaning, exploratory analysis, hypothesis testing, statistical reports, ML model evaluation, dataset profiling, charts, dashboards. Upload CSVs and get analysis back with full Python access.\"\nmetadata:\n openclaw:\n emoji: \"π’\"\n os: [darwin, linux, windows]\n requires:\n bins: [python3]\n env: [CELLCOG_API_KEY]\nauthor: CellCog\nhomepage: https://cellcog.ai\ndependencies: [cellcog]\n---\n# Data Analysis - Your Data Has Answers, CellCog Finds Them\n\nData analysis and visualization from uploaded files.\n\nMost AI tools return code when you ask about data. CellCog returns answers β actual charts, clean datasets, statistical reports, and visual dashboards. Upload messy CSVs with a minimal prompt, and CellCog's coding agent explores your data, finds the patterns, and presents them beautifully. Full Python access for everything from data cleaning to ML model evaluation.\n\n## How to Use\n\nFor your first CellCog task in a session, read the **cellcog** skill for the full SDK reference β file handling, chat modes, timeouts, and more.\n\n**OpenClaw (fire-and-forget):**\n```python\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n notify_session_key=\"agent:main:main\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n chat_tier=\"max\",\n)\n```\n\n**All agents except OpenClaw (blocks until done):**\n```python\nfrom cellcog import CellCogClient\nclient = CellCogClient(agent_provider=\"openclaw|cursor|claude-code|codex|...\")\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n chat_tier=\"max\",\n)\nprint(result[\"message\"])\n```\n\n\n---\n\n## What Makes Data-Cog Different\n\n### Code as Tool, Not as Output\n\nOther AI tools give you Python code and say \"run this.\" CellCog **runs the code for you** and delivers the results:\n\n| Other AI Tools | Data-Cog |\n|---------------|----------|\n| \"Here's a pandas script to analyze your data\" | Here are your actual insights with charts |\n| \"Run this matplotlib code to see the chart\" | Here's the chart, annotated with findings |\n| \"This SQL query will find outliers\" | Found 23 outliers, here's what they mean |\n| \"You'll need scikit-learn for this\" | Model trained, here's accuracy and feature importance |\n\nYou upload data. You get answers. The code runs behind the scenes.\n\n---\n\n## What Data Work You Can Do\n\n### Exploratory Data Analysis\n\nUnderstand your data fast:\n\n- **Dataset Profiling**: \"Analyze this CSV β distributions, missing values, outliers, correlations, and data quality summary\"\n- **Pattern Discovery**: \"What patterns and trends exist in this sales data? Surprise me.\"\n- **Anomaly Detection**: \"Find unusual patterns in this server log data β what looks abnormal?\"\n- **Relationship Analysis**: \"What factors most strongly correlate with customer churn in this dataset?\"\n\n**Example prompt:**\n> \"Analyze this dataset:\n> /path/to/customer_data.csv\n> \n> I don't know much about this data yet. Give me:\n> - Overview: rows, columns, data types, missing values\n> - Key distributions and summary statistics\n> - Most interesting correlations\n> - Any outliers or data quality issues\n> - 3-5 insights that jump out\n> \n> Present findings as an interactive HTML report with charts.\"\n\n### Data Cleaning & Transformation\n\nWrangle messy data into shape:\n\n- **Clean Messy Data**: \"Clean this CSV β fix inconsistent date formats, handle missing values, remove duplicates, standardize column names\"\n- **Data Transformation**: \"Pivot this transaction data into a monthly summary by product category\"\n- **Data Merging**: \"Join these three CSV files on customer_id and create a unified dataset\"\n- **Feature Engineering**: \"Create useful features from this raw data for predicting house prices\"\n\n**Example prompt:**\n> \"Clean and transform this dataset:\n> /path/to/messy_data.csv\n> \n> Issues I know about:\n> - Dates are in mixed formats (MM/DD/YYYY and YYYY-MM-DD)\n> - 'Revenue' column has some values with $ signs and commas\n> - Duplicate rows exist\n> - Missing values in 'Region' column\n> \n> Clean it up and give me back a clean CSV plus a summary of what you changed.\"\n\n### Statistical Analysis\n\nRigorous analysis with real numbers:\n\n- **Hypothesis Testing**: \"Is there a statistically significant difference in conversion rates between our A and B variants?\"\n- **Regression Analysis**: \"What factors predict employee salary in this HR dataset? Build a regression model.\"\n- **Time Series Analysis**: \"Analyze this monthly revenue data β trend, seasonality, and forecast next 6 months\"\n- **Cohort Analysis**: \"Create a cohort analysis showing user retention by signup month\"\n\n**Example prompt:**\n> \"I ran an A/B test on our checkout page:\n> /path/to/ab_test_results.csv\n> \n> Columns: user_id, variant (A or B), converted (0/1), revenue, timestamp\n> \n> Tell me:\n> - Is variant B statistically better? (p-value, confidence interval)\n> - Conversion rate difference\n> - Revenue per user difference\n> - Sample size adequacy check\n> - My recommendation: ship B or keep testing?\n> \n> Present with clear charts and a plain-English conclusion.\"\n\n### Visualization & Reporting\n\nTurn data into visual stories:\n\n- **Chart Generation**: \"Create a set of charts showing our quarterly performance from this data\"\n- **Dashboard Reports**: \"Build an interactive dashboard from this sales dataset with filters by region and product\"\n- **Presentation-Ready Visuals**: \"Create publication-quality charts from this research data\"\n- **Comparison Visuals**: \"Visualize how our metrics compare to industry benchmarks\"\n\n### Machine Learning\n\nApplied ML without the setup:\n\n- **Classification**: \"Predict which customers will churn based on this dataset β train a model, show feature importance\"\n- **Clustering**: \"Segment these customers into groups based on behavior β how many natural clusters exist?\"\n- **Forecasting**: \"Forecast next quarter's sales using this historical data\"\n- **Model Evaluation**: \"I trained a model β here are the predictions. Evaluate: accuracy, precision, recall, confusion matrix, ROC curve\"\n\n**Example prompt:**\n> \"Predict customer churn from this dataset:\n> /path/to/customer_features.csv\n> \n> Target column: 'churned'\n> \n> - Train a model, try at least 2 algorithms\n> - Show feature importance β what drives churn?\n> - Confusion matrix and ROC curve\n> - Plain-English summary: 'The top 3 reasons customers churn are...'\n> - Actionable recommendations based on findings\n> \n> I want insights, not just metrics.\"\n\n---\n\n## Supported Data Formats\n\n| Format | How to Send |\n|--------|-------------|\n| **CSV** | Upload via SHOW_FILE |\n| **Excel (XLSX)** | Upload via SHOW_FILE |\n| **JSON** | Upload via SHOW_FILE |\n| **Parquet** | Upload via SHOW_FILE |\n| **SQL exports** | Upload the dump via SHOW_FILE |\n| **Inline data** | Describe small datasets directly in prompt |\n\n---\n\n## Output Formats\n\n| Format | Best For |\n|--------|----------|\n| **Interactive HTML Dashboard** | Explorable charts, filters, drill-downs |\n| **PDF Report** | Shareable analysis reports with charts and findings |\n| **Clean CSV/XLSX** | Cleaned or transformed data files for downstream use |\n| **Markdown** | Quick insights for integration into docs |\n\n---\n\n## Choosing Mode & Tier\n\n**Use `chat_mode=\"agent\", chat_tier=\"max\"` for data analysis.** Analysis is coding work β pipelines, statistics, and modeling need the deepest tier.\n\n| Scenario | Recommended |\n|----------|-------------|\n| All data analysis and ML work | `chat_mode=\"agent\", chat_tier=\"max\"` |\n| Trivial one-column summaries | `chat_mode=\"agent\"` (defaults to `\"flash\"`) |\n\nAgent Team (`chat_mode=\"team\"`) is reserved for deep research β data analysis runs best on Agent max.\n\n---\n\n## Example Prompts\n\n**Minimal prompt, maximum insight:**\n> \"Analyze this:\n> /path/to/data.csv\n> \n> Tell me everything interesting.\"\n\nThat's it. CellCog's coding agent will profile the data, run exploratory analysis, find patterns, and present findings with charts. You don't need to know what to ask β the agent figures it out.\n\n**Business analysis:**\n> \"Analyze our e-commerce data:\n> /path/to/orders.csv\n> \n> I need:\n> - Revenue trends (daily, weekly, monthly)\n> - Best and worst performing products\n> - Customer purchase frequency distribution\n> - Average order value trends\n> - Seasonal patterns\n> - Top 5 actionable insights for growing revenue\n> \n> Interactive HTML dashboard with all charts.\"\n\n**Research data analysis:**\n> \"Analyze this survey data from 500 respondents:\n> /path/to/survey.csv\n> \n> Research questions:\n> 1. Is there a significant relationship between age group and product preference?\n> 2. Do satisfaction scores differ by region? (ANOVA)\n> 3. What factors best predict likelihood to recommend? (regression)\n> \n> Include: statistical tests, p-values, effect sizes, and publication-ready charts.\n> PDF report format.\"\n\n---\n\n## Tips for Better Data Analysis\n\n1. **Just upload and ask**: You don't need to describe every column. CellCog reads the data and figures out what's there.\n\n2. **State your question**: \"What drives churn?\" is more focused than \"Analyze this data.\" Both work, but the first gets faster results.\n\n3. **Mention the audience**: \"For my CEO\" means executive summary. \"For the data team\" means show the methodology.\n\n4. **Specify what you'll do with it**: \"I need to present this to the board\" vs \"I need clean data for my ML pipeline\" β context shapes the output.\n\n5. **Don't over-specify methods**: Let CellCog choose the right statistical approach. Say what you want to *learn*, not which algorithm to use.\n\n6. **Iterate**: Upload data β get initial analysis β ask follow-up questions β go deeper. CellCog maintains context across messages.\n\n---\n\n## If CellCog is not installed\n\n**Claude Code, Cursor, Codex + 70 more agents:** `npx skills add cellcog/skills --skill cellcog`\n**OpenClaw:** `openclaw skills install @cellcog/cellcog`\n**CellCog plugin users:** run `/cellcog-setup` (or `/cellcog:cellcog-setup` depending on your tool)\n**Manual setup:** `pip install -U cellcog` and set `CELLCOG_API_KEY`. See the **cellcog** skill for SDK reference.\n\nFile v1.0.16:_meta.json\n\n{\n \"ownerId\": \"kn7a96cj9q65e0bhmzahv790en80ffqm\",\n \"slug\": \"data-analysis-cellcog\",\n \"version\": \"1.0.16\",\n \"publishedAt\": 1787536870439\n}\n\nFile v1.0.16:skill-card.md\n\n## Description:\n\nAI data analysis and visualization powered by CellCog. Data cleaning, exploratory analysis, hypothesis testing, statistical reports, ML model evaluation, dataset profiling, charts, dashboards.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[cellcog](https://clawhub.ai/user/cellcog)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers, analysts, and other external users use this skill to send datasets to CellCog for cleaning, exploratory analysis, statistical testing, visualization, reporting, and machine learning evaluation.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Selected datasets are uploaded to CellCog, a third-party analysis service.\n\nMitigation: Use only datasets your organization permits for third-party processing, and redact or minimize secrets, credentials, regulated records, and confidential business data before upload.\n\nRisk: Setup guidance may use latest-version dependency installation commands.\n\nMitigation: Pin or verify dependency versions where reproducibility or supply-chain review is required.\n\n## Reference(s):\n\n- [CellCog homepage](https://cellcog.ai)\n- [ClawHub skill page](https://clawhub.ai/cellcog/skills/data-analysis-cellcog)\n- [CellCog publisher profile](https://clawhub.ai/user/cellcog)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, code, shell commands, configuration, guidance, files]\n\n**Output Format:** [Markdown guidance with Python examples, API call patterns, and references to generated analysis artifacts such as charts, dashboards, reports, and cleaned datasets.]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Uses uploaded datasets and a CellCog API key; outputs may include HTML dashboards, PDF reports, CSV or XLSX files, and markdown summaries depending on the user request.]\n\n## Skill Version(s):\n\n1.0.16 (source: ClawHub release evidence)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nArchive v1.0.15: 3 files, 5945 bytes\n\nFiles: skill-card.md (2088b), SKILL.md (10314b), _meta.json (141b)\n\nFile v1.0.15:SKILL.md\n\n---\nname: data-analysis-cellcog\ndescription: \"AI data analysis and visualization powered by CellCog. Data cleaning, exploratory analysis, hypothesis testing, statistical reports, ML model evaluation, dataset profiling, charts, dashboards. Upload CSVs and get analysis back with full Python access.\"\nmetadata:\n openclaw:\n emoji: \"π’\"\n os: [darwin, linux, windows]\n requires:\n bins: [python3]\n env: [CELLCOG_API_KEY]\nauthor: CellCog\nhomepage: https://cellcog.ai\ndependencies: [cellcog]\n---\n# Data Analysis - Your Data Has Answers, CellCog Finds Them\n\nData analysis and visualization from uploaded files.\n\nMost AI tools return code when you ask about data. CellCog returns answers β actual charts, clean datasets, statistical reports, and visual dashboards. Upload messy CSVs with a minimal prompt, and CellCog's coding agent explores your data, finds the patterns, and presents them beautifully. Full Python access for everything from data cleaning to ML model evaluation.\n\n## How to Use\n\nFor your first CellCog task in a session, read the **cellcog** skill for the full SDK reference β file handling, chat modes, timeouts, and more.\n\n**OpenClaw (fire-and-forget):**\n```python\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n notify_session_key=\"agent:main:main\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n chat_tier=\"max\",\n)\n```\n\n**All agents except OpenClaw (blocks until done):**\n```python\nfrom cellcog import CellCogClient\nclient = CellCogClient(agent_provider=\"openclaw|cursor|claude-code|codex|...\")\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n chat_tier=\"max\",\n)\nprint(result[\"message\"])\n```\n\n\n---\n\n## What Makes Data-Cog Different\n\n### Code as Tool, Not as Output\n\nOther AI tools give you Python code and say \"run this.\" CellCog **runs the code for you** and delivers the results:\n\n| Other AI Tools | Data-Cog |\n|---------------|----------|\n| \"Here's a pandas script to analyze your data\" | Here are your actual insights with charts |\n| \"Run this matplotlib code to see the chart\" | Here's the chart, annotated with findings |\n| \"This SQL query will find outliers\" | Found 23 outliers, here's what they mean |\n| \"You'll need scikit-learn for this\" | Model trained, here's accuracy and feature importance |\n\nYou upload data. You get answers. The code runs behind the scenes.\n\n---\n\n## What Data Work You Can Do\n\n### Exploratory Data Analysis\n\nUnderstand your data fast:\n\n- **Dataset Profiling**: \"Analyze this CSV β distributions, missing values, outliers, correlations, and data quality summary\"\n- **Pattern Discovery**: \"What patterns and trends exist in this sales data? Surprise me.\"\n- **Anomaly Detection**: \"Find unusual patterns in this server log data β what looks abnormal?\"\n- **Relationship Analysis**: \"What factors most strongly correlate with customer churn in this dataset?\"\n\n**Example prompt:**\n> \"Analyze this dataset:\n> /path/to/customer_data.csv\n> \n> I don't know much about this data yet. Give me:\n> - Overview: rows, columns, data types, missing values\n> - Key distributions and summary statistics\n> - Most interesting correlations\n> - Any outliers or data quality issues\n> - 3-5 insights that jump out\n> \n> Present findings as an interactive HTML report with charts.\"\n\n### Data Cleaning & Transformation\n\nWrangle messy data into shape:\n\n- **Clean Messy Data**: \"Clean this CSV β fix inconsistent date formats, handle missing values, remove duplicates, standardize column names\"\n- **Data Transformation**: \"Pivot this transaction data into a monthly summary by product category\"\n- **Data Merging**: \"Join these three CSV files on customer_id and create a unified dataset\"\n- **Feature Engineering**: \"Create useful features from this raw data for predicting house prices\"\n\n**Example prompt:**\n> \"Clean and transform this dataset:\n> /path/to/messy_data.csv\n> \n> Issues I know about:\n> - Dates are in mixed formats (MM/DD/YYYY and YYYY-MM-DD)\n> - 'Revenue' column has some values with $ signs and commas\n> - Duplicate rows exist\n> - Missing values in 'Region' column\n> \n> Clean it up and give me back a clean CSV plus a summary of what you changed.\"\n\n### Statistical Analysis\n\nRigorous analysis with real numbers:\n\n- **Hypothesis Testing**: \"Is there a statistically significant difference in conversion rates between our A and B variants?\"\n- **Regression Analysis**: \"What factors predict employee salary in this HR dataset? Build a regression model.\"\n- **Time Series Analysis**: \"Analyze this monthly revenue data β trend, seasonality, and forecast next 6 months\"\n- **Cohort Analysis**: \"Create a cohort analysis showing user retention by signup month\"\n\n**Example prompt:**\n> \"I ran an A/B test on our checkout page:\n> /path/to/ab_test_results.csv\n> \n> Columns: user_id, variant (A or B), converted (0/1), revenue, timestamp\n> \n> Tell me:\n> - Is variant B statistically better? (p-value, confidence interval)\n> - Conversion rate difference\n> - Revenue per user difference\n> - Sample size adequacy check\n> - My recommendation: ship B or keep testing?\n> \n> Present with clear charts and a plain-English conclusion.\"\n\n### Visualization & Reporting\n\nTurn data into visual stories:\n\n- **Chart Generation**: \"Create a set of charts showing our quarterly performance from this data\"\n- **Dashboard Reports**: \"Build an interactive dashboard from this sales dataset with filters by region and product\"\n- **Presentation-Ready Visuals**: \"Create publication-quality charts from this research data\"\n- **Comparison Visuals**: \"Visualize how our metrics compare to industry benchmarks\"\n\n### Machine Learning\n\nApplied ML without the setup:\n\n- **Classification**: \"Predict which customers will churn based on this dataset β train a model, show feature importance\"\n- **Clustering**: \"Segment these customers into groups based on behavior β how many natural clusters exist?\"\n- **Forecasting**: \"Forecast next quarter's sales using this historical data\"\n- **Model Evaluation**: \"I trained a model β here are the predictions. Evaluate: accuracy, precision, recall, confusion matrix, ROC curve\"\n\n**Example prompt:**\n> \"Predict customer churn from this dataset:\n> /path/to/customer_features.csv\n> \n> Target column: 'churned'\n> \n> - Train a model, try at least 2 algorithms\n> - Show feature importance β what drives churn?\n> - Confusion matrix and ROC curve\n> - Plain-English summary: 'The top 3 reasons customers churn are...'\n> - Actionable recommendations based on findings\n> \n> I want insights, not just metrics.\"\n\n---\n\n## Supported Data Formats\n\n| Format | How to Send |\n|--------|-------------|\n| **CSV** | Upload via SHOW_FILE |\n| **Excel (XLSX)** | Upload via SHOW_FILE |\n| **JSON** | Upload via SHOW_FILE |\n| **Parquet** | Upload via SHOW_FILE |\n| **SQL exports** | Upload the dump via SHOW_FILE |\n| **Inline data** | Describe small datasets directly in prompt |\n\n---\n\n## Output Formats\n\n| Format | Best For |\n|--------|----------|\n| **Interactive HTML Dashboard** | Explorable charts, filters, drill-downs |\n| **PDF Report** | Shareable analysis reports with charts and findings |\n| **Clean CSV/XLSX** | Cleaned or transformed data files for downstream use |\n| **Markdown** | Quick insights for integration into docs |\n\n---\n\n## Choosing Mode & Tier\n\n**Use `chat_mode=\"agent\", chat_tier=\"max\"` for data analysis.** Analysis is coding work β pipelines, statistics, and modeling need the deepest tier.\n\n| Scenario | Recommended |\n|----------|-------------|\n| All data analysis and ML work | `chat_mode=\"agent\", chat_tier=\"max\"` |\n| Trivial one-column summaries | `chat_mode=\"agent\"` (defaults to `\"flash\"`) |\n\nAgent Team (`chat_mode=\"team\"`) is reserved for deep research β data analysis runs best on Agent max.\n\n---\n\n## Example Prompts\n\n**Minimal prompt, maximum insight:**\n> \"Analyze this:\n> /path/to/data.csv\n> \n> Tell me everything interesting.\"\n\nThat's it. CellCog's coding agent will profile the data, run exploratory analysis, find patterns, and present findings with charts. You don't need to know what to ask β the agent figures it out.\n\n**Business analysis:**\n> \"Analyze our e-commerce data:\n> /path/to/orders.csv\n> \n> I need:\n> - Revenue trends (daily, weekly, monthly)\n> - Best and worst performing products\n> - Customer purchase frequency distribution\n> - Average order value trends\n> - Seasonal patterns\n> - Top 5 actionable insights for growing revenue\n> \n> Interactive HTML dashboard with all charts.\"\n\n**Research data analysis:**\n> \"Analyze this survey data from 500 respondents:\n> /path/to/survey.csv\n> \n> Research questions:\n> 1. Is there a significant relationship between age group and product preference?\n> 2. Do satisfaction scores differ by region? (ANOVA)\n> 3. What factors best predict likelihood to recommend? (regression)\n> \n> Include: statistical tests, p-values, effect sizes, and publication-ready charts.\n> PDF report format.\"\n\n---\n\n## Tips for Better Data Analysis\n\n1. **Just upload and ask**: You don't need to describe every column. CellCog reads the data and figures out what's there.\n\n2. **State your question**: \"What drives churn?\" is more focused than \"Analyze this data.\" Both work, but the first gets faster results.\n\n3. **Mention the audience**: \"For my CEO\" means executive summary. \"For the data team\" means show the methodology.\n\n4. **Specify what you'll do with it**: \"I need to present this to the board\" vs \"I need clean data for my ML pipeline\" β context shapes the output.\n\n5. **Don't over-specify methods**: Let CellCog choose the right statistical approach. Say what you want to *learn*, not which algorithm to use.\n\n6. **Iterate**: Upload data β get initial analysis β ask follow-up questions β go deeper. CellCog maintains context across messages.\n\n---\n\n## If CellCog is not installed\n\n**Claude Code, Cursor, Codex + 70 more agents:** `npx skills add cellcog/skills --skill cellcog`\n**OpenClaw:** `clawhub install cellcog`\n**CellCog plugin users:** run `/cellcog-setup` (or `/cellcog:cellcog-setup` depending on your tool)\n**Manual setup:** `pip install -U cellcog` and set `CELLCOG_API_KEY`. See the **cellcog** skill for SDK reference.\n\nFile v1.0.15:_meta.json\n\n{\n \"ownerId\": \"kn7a96cj9q65e0bhmzahv790en80ffqm\",\n \"slug\": \"data-analysis-cellcog\",\n \"version\": \"1.0.15\",\n \"publishedAt\": 1787535928588\n}\n\nFile v1.0.15:skill-card.md\n\n## Description:\n\nCellCog's Data Analysis skill helps agents analyze uploaded datasets, produce charts and reports, clean and transform data, run statistical analysis, and evaluate machine learning results.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[cellcog](https://clawhub.ai/user/cellcog)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nExternal users and developers use this skill to send datasets to CellCog for exploratory analysis, data cleaning, statistical testing, visualization, dashboarding, and machine learning evaluation. It is suited to workflows where the agent should return findings, reports, charts, or cleaned data rather than only suggesting code.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Uploaded datasets are processed by an external CellCog service.\n\nMitigation: Confirm the dataset is approved for use with CellCog, and do not upload secrets, credentials, regulated personal data, customer records, HR data, proprietary logs, or confidential research unless the organization's privacy, retention, and security requirements are satisfied.\n\n## Reference(s):\n\n- [ClawHub skill page](https://clawhub.ai/cellcog/skills/data-analysis-cellcog)\n- [CellCog homepage](https://cellcog.ai)\n- [CellCog publisher profile](https://clawhub.ai/user/cellcog)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, code, files, configuration, guidance]\n\n**Output Format:** [Markdown or text responses with optional generated reports, charts, dashboards, and cleaned data files.]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May return interactive HTML dashboards, PDF reports, CSV/XLSX files, Markdown summaries, charts, and plain-language analysis.]\n\n## Skill Version(s):\n\n1.0.15 (source: server release metadata)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nArchive v1.0.14: 3 files, 5942 bytes\n\nFiles: skill-card.md (2125b), SKILL.md (10349b), _meta.json (141b)\n\nFile v1.0.14:SKILL.md\n\n---\nname: data-analysis-cellcog\ndescription: \"AI data analysis and visualization powered by CellCog. Data cleaning, exploratory analysis, hypothesis testing, statistical reports, ML model evaluation, dataset profiling, charts, dashboards. Upload CSVs and get analysis back with full Python access.\"\nmetadata:\n openclaw:\n emoji: \"π’\"\n os: [darwin, linux, windows]\n requires:\n bins: [python3]\n env: [CELLCOG_API_KEY]\nauthor: CellCog\nhomepage: https://cellcog.ai\ndependencies: [cellcog]\n---\n# Data Analysis - Your Data Has Answers, CellCog Finds Them\n\nData analysis and visualization from uploaded files.\n\nMost AI tools return code when you ask about data. CellCog returns answers β actual charts, clean datasets, statistical reports, and visual dashboards. Upload messy CSVs with a minimal prompt, and CellCog's coding agent explores your data, finds the patterns, and presents them beautifully. Full Python access for everything from data cleaning to ML model evaluation.\n\n## How to Use\n\nFor your first CellCog task in a session, read the **cellcog** skill for the full SDK reference β file handling, chat modes, timeouts, and more.\n\n**OpenClaw (fire-and-forget):**\n```python\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n notify_session_key=\"agent:main:main\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n)\n```\n\n**All agents except OpenClaw (blocks until done):**\n```python\nfrom cellcog import CellCogClient\nclient = CellCogClient(agent_provider=\"openclaw|cursor|claude-code|codex|...\")\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n)\nprint(result[\"message\"])\n```\n\n\n---\n\n## What Makes Data-Cog Different\n\n### Code as Tool, Not as Output\n\nOther AI tools give you Python code and say \"run this.\" CellCog **runs the code for you** and delivers the results:\n\n| Other AI Tools | Data-Cog |\n|---------------|----------|\n| \"Here's a pandas script to analyze your data\" | Here are your actual insights with charts |\n| \"Run this matplotlib code to see the chart\" | Here's the chart, annotated with findings |\n| \"This SQL query will find outliers\" | Found 23 outliers, here's what they mean |\n| \"You'll need scikit-learn for this\" | Model trained, here's accuracy and feature importance |\n\nYou upload data. You get answers. The code runs behind the scenes.\n\n---\n\n## What Data Work You Can Do\n\n### Exploratory Data Analysis\n\nUnderstand your data fast:\n\n- **Dataset Profiling**: \"Analyze this CSV β distributions, missing values, outliers, correlations, and data quality summary\"\n- **Pattern Discovery**: \"What patterns and trends exist in this sales data? Surprise me.\"\n- **Anomaly Detection**: \"Find unusual patterns in this server log data β what looks abnormal?\"\n- **Relationship Analysis**: \"What factors most strongly correlate with customer churn in this dataset?\"\n\n**Example prompt:**\n> \"Analyze this dataset:\n> /path/to/customer_data.csv\n> \n> I don't know much about this data yet. Give me:\n> - Overview: rows, columns, data types, missing values\n> - Key distributions and summary statistics\n> - Most interesting correlations\n> - Any outliers or data quality issues\n> - 3-5 insights that jump out\n> \n> Present findings as an interactive HTML report with charts.\"\n\n### Data Cleaning & Transformation\n\nWrangle messy data into shape:\n\n- **Clean Messy Data**: \"Clean this CSV β fix inconsistent date formats, handle missing values, remove duplicates, standardize column names\"\n- **Data Transformation**: \"Pivot this transaction data into a monthly summary by product category\"\n- **Data Merging**: \"Join these three CSV files on customer_id and create a unified dataset\"\n- **Feature Engineering**: \"Create useful features from this raw data for predicting house prices\"\n\n**Example prompt:**\n> \"Clean and transform this dataset:\n> /path/to/messy_data.csv\n> \n> Issues I know about:\n> - Dates are in mixed formats (MM/DD/YYYY and YYYY-MM-DD)\n> - 'Revenue' column has some values with $ signs and commas\n> - Duplicate rows exist\n> - Missing values in 'Region' column\n> \n> Clean it up and give me back a clean CSV plus a summary of what you changed.\"\n\n### Statistical Analysis\n\nRigorous analysis with real numbers:\n\n- **Hypothesis Testing**: \"Is there a statistically significant difference in conversion rates between our A and B variants?\"\n- **Regression Analysis**: \"What factors predict employee salary in this HR dataset? Build a regression model.\"\n- **Time Series Analysis**: \"Analyze this monthly revenue data β trend, seasonality, and forecast next 6 months\"\n- **Cohort Analysis**: \"Create a cohort analysis showing user retention by signup month\"\n\n**Example prompt:**\n> \"I ran an A/B test on our checkout page:\n> /path/to/ab_test_results.csv\n> \n> Columns: user_id, variant (A or B), converted (0/1), revenue, timestamp\n> \n> Tell me:\n> - Is variant B statistically better? (p-value, confidence interval)\n> - Conversion rate difference\n> - Revenue per user difference\n> - Sample size adequacy check\n> - My recommendation: ship B or keep testing?\n> \n> Present with clear charts and a plain-English conclusion.\"\n\n### Visualization & Reporting\n\nTurn data into visual stories:\n\n- **Chart Generation**: \"Create a set of charts showing our quarterly performance from this data\"\n- **Dashboard Reports**: \"Build an interactive dashboard from this sales dataset with filters by region and product\"\n- **Presentation-Ready Visuals**: \"Create publication-quality charts from this research data\"\n- **Comparison Visuals**: \"Visualize how our metrics compare to industry benchmarks\"\n\n### Machine Learning\n\nApplied ML without the setup:\n\n- **Classification**: \"Predict which customers will churn based on this dataset β train a model, show feature importance\"\n- **Clustering**: \"Segment these customers into groups based on behavior β how many natural clusters exist?\"\n- **Forecasting**: \"Forecast next quarter's sales using this historical data\"\n- **Model Evaluation**: \"I trained a model β here are the predictions. Evaluate: accuracy, precision, recall, confusion matrix, ROC curve\"\n\n**Example prompt:**\n> \"Predict customer churn from this dataset:\n> /path/to/customer_features.csv\n> \n> Target column: 'churned'\n> \n> - Train a model, try at least 2 algorithms\n> - Show feature importance β what drives churn?\n> - Confusion matrix and ROC curve\n> - Plain-English summary: 'The top 3 reasons customers churn are...'\n> - Actionable recommendations based on findings\n> \n> I want insights, not just metrics.\"\n\n---\n\n## Supported Data Formats\n\n| Format | How to Send |\n|--------|-------------|\n| **CSV** | Upload via SHOW_FILE |\n| **Excel (XLSX)** | Upload via SHOW_FILE |\n| **JSON** | Upload via SHOW_FILE |\n| **Parquet** | Upload via SHOW_FILE |\n| **SQL exports** | Upload the dump via SHOW_FILE |\n| **Inline data** | Describe small datasets directly in prompt |\n\n---\n\n## Output Formats\n\n| Format | Best For |\n|--------|----------|\n| **Interactive HTML Dashboard** | Explorable charts, filters, drill-downs |\n| **PDF Report** | Shareable analysis reports with charts and findings |\n| **Clean CSV/XLSX** | Cleaned or transformed data files for downstream use |\n| **Markdown** | Quick insights for integration into docs |\n\n---\n\n## Chat Mode for Data\n\n| Scenario | Recommended Mode |\n|----------|------------------|\n| Quick data cleaning, simple charts, basic statistics | `\"agent\"` |\n| Deep analysis with multiple techniques, ML modeling, comprehensive reports | `\"agent team\"` |\n\n**Use `\"agent\"` for most data work.** Data cleaning, EDA, chart generation, and standard statistical analysis execute well in agent mode.\n\n**Use `\"agent team\"` for complex analytical projects** β multi-technique analysis, ML model comparisons, or when you need deep domain reasoning about what the data means.\n\n---\n\n## Example Prompts\n\n**Minimal prompt, maximum insight:**\n> \"Analyze this:\n> /path/to/data.csv\n> \n> Tell me everything interesting.\"\n\nThat's it. CellCog's coding agent will profile the data, run exploratory analysis, find patterns, and present findings with charts. You don't need to know what to ask β the agent figures it out.\n\n**Business analysis:**\n> \"Analyze our e-commerce data:\n> /path/to/orders.csv\n> \n> I need:\n> - Revenue trends (daily, weekly, monthly)\n> - Best and worst performing products\n> - Customer purchase frequency distribution\n> - Average order value trends\n> - Seasonal patterns\n> - Top 5 actionable insights for growing revenue\n> \n> Interactive HTML dashboard with all charts.\"\n\n**Research data analysis:**\n> \"Analyze this survey data from 500 respondents:\n> /path/to/survey.csv\n> \n> Research questions:\n> 1. Is there a significant relationship between age group and product preference?\n> 2. Do satisfaction scores differ by region? (ANOVA)\n> 3. What factors best predict likelihood to recommend? (regression)\n> \n> Include: statistical tests, p-values, effect sizes, and publication-ready charts.\n> PDF report format.\"\n\n---\n\n## Tips for Better Data Analysis\n\n1. **Just upload and ask**: You don't need to describe every column. CellCog reads the data and figures out what's there.\n\n2. **State your question**: \"What drives churn?\" is more focused than \"Analyze this data.\" Both work, but the first gets faster results.\n\n3. **Mention the audience**: \"For my CEO\" means executive summary. \"For the data team\" means show the methodology.\n\n4. **Specify what you'll do with it**: \"I need to present this to the board\" vs \"I need clean data for my ML pipeline\" β context shapes the output.\n\n5. **Don't over-specify methods**: Let CellCog choose the right statistical approach. Say what you want to *learn*, not which algorithm to use.\n\n6. **Iterate**: Upload data β get initial analysis β ask follow-up questions β go deeper. CellCog maintains context across messages.\n\n---\n\n## If CellCog is not installed\n\n**Claude Code, Cursor, Codex + 70 more agents:** `npx skills add cellcog/skills --skill cellcog`\n**OpenClaw:** `clawhub install cellcog`\n**CellCog plugin users:** run `/cellcog-setup` (or `/cellcog:cellcog-setup` depending on your tool)\n**Manual setup:** `pip install -U cellcog` and set `CELLCOG_API_KEY`. See the **cellcog** skill for SDK reference.\n\nFile v1.0.14:_meta.json\n\n{\n \"ownerId\": \"kn7a96cj9q65e0bhmzahv790en80ffqm\",\n \"slug\": \"data-analysis-cellcog\",\n \"version\": \"1.0.14\",\n \"publishedAt\": 1785737152041\n}\n\nFile v1.0.14:skill-card.md\n\n## Description:
\nAI data analysis and visualization powered by CellCog. Data cleaning, exploratory analysis, hypothesis testing, statistical reports, ML model evaluation, dataset profiling, charts, dashboards. Upload CSVs and get analysis back with full Python access.
\n\nThis skill is ready for commercial/non-commercial use.
\n\n## Publisher:
\n[cellcog](https://clawhub.ai/user/cellcog)
\n\n### License/Terms of Use:
\nMIT-0
\n\n\n## Use Case:
\nDevelopers, analysts, and business users use this skill to send datasets to CellCog for data cleaning, exploratory analysis, statistical testing, visualization, reporting, and ML model evaluation.
\n\n### Deployment Geography for Use:
\nGlobal
\n\n## Known Risks and Mitigations:
\nRisk: Uploaded analysis files may be sent to CellCog for processing.
\nMitigation: Avoid regulated data, secrets, credentials, and confidential business datasets unless organizational policy permits CellCog for that use.
\nRisk: The skill requires CELLCOG_API_KEY to access the CellCog service.
\nMitigation: Store the API key in the environment and do not include credentials in prompts, uploaded datasets, reports, or generated artifacts.
\n\n\n## Reference(s):
\n- [ClawHub skill page](https://clawhub.ai/cellcog/skills/data-analysis-cellcog)
\n- [CellCog homepage](https://cellcog.ai)
\n\n\n## Skill Output:
\n**Output Type(s):** [text, markdown, code, shell commands, configuration, guidance, files]
\n**Output Format:** [Markdown, Python snippets, generated analysis files, charts, dashboards, reports, and cleaned datasets]
\n**Output Parameters:** [1D]
\n**Other Properties Related to Output:** [Requires python3 and CELLCOG_API_KEY; supports darwin, linux, and windows.]
\n\n## Skill Version(s):
\n1.0.14 (source: server release evidence)
\n\n## Ethical Considerations:
\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.
\n\nArchive v1.0.13: 3 files, 5854 bytes\n\nFiles: skill-card.md (1991b), SKILL.md (10274b), _meta.json (141b)\n\nFile v1.0.13:SKILL.md\n\n---\nname: data-analysis-cellcog\ndescription: \"AI data analysis and visualization powered by CellCog. Data cleaning, exploratory analysis, hypothesis testing, statistical reports, ML model evaluation, dataset profiling, charts, dashboards. Upload CSVs and get analysis back with full Python access.\"\nmetadata:\n openclaw:\n emoji: \"π’\"\n os: [darwin, linux, windows]\n requires:\n bins: [python3]\n env: [CELLCOG_API_KEY]\nauthor: CellCog\nhomepage: https://cellcog.ai\ndependencies: [cellcog]\n---\n# Data Analysis - Your Data Has Answers, CellCog Finds Them\n\nData analysis and visualization from uploaded files.\n\nMost AI tools return code when you ask about data. CellCog returns answers β actual charts, clean datasets, statistical reports, and visual dashboards. Upload messy CSVs with a minimal prompt, and CellCog's coding agent explores your data, finds the patterns, and presents them beautifully. Full Python access for everything from data cleaning to ML model evaluation.\n\n## How to Use\n\nFor your first CellCog task in a session, read the **cellcog** skill for the full SDK reference β file handling, chat modes, timeouts, and more.\n\n**OpenClaw (fire-and-forget):**\n```python\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n notify_session_key=\"agent:main:main\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n)\n```\n\n**All agents except OpenClaw (blocks until done):**\n```python\nfrom cellcog import CellCogClient\nclient = CellCogClient(agent_provider=\"openclaw|cursor|claude-code|codex|...\")\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n)\nprint(result[\"message\"])\n```\n\n\n---\n\n## What Makes Data-Cog Different\n\n### Code as Tool, Not as Output\n\nOther AI tools give you Python code and say \"run this.\" CellCog **runs the code for you** and delivers the results:\n\n| Other AI Tools | Data-Cog |\n|---------------|----------|\n| \"Here's a pandas script to analyze your data\" | Here are your actual insights with charts |\n| \"Run this matplotlib code to see the chart\" | Here's the chart, annotated with findings |\n| \"This SQL query will find outliers\" | Found 23 outliers, here's what they mean |\n| \"You'll need scikit-learn for this\" | Model trained, here's accuracy and feature importance |\n\nYou upload data. You get answers. The code runs behind the scenes.\n\n---\n\n## What Data Work You Can Do\n\n### Exploratory Data Analysis\n\nUnderstand your data fast:\n\n- **Dataset Profiling**: \"Analyze this CSV β distributions, missing values, outliers, correlations, and data quality summary\"\n- **Pattern Discovery**: \"What patterns and trends exist in this sales data? Surprise me.\"\n- **Anomaly Detection**: \"Find unusual patterns in this server log data β what looks abnormal?\"\n- **Relationship Analysis**: \"What factors most strongly correlate with customer churn in this dataset?\"\n\n**Example prompt:**\n> \"Analyze this dataset:\n> /path/to/customer_data.csv\n> \n> I don't know much about this data yet. Give me:\n> - Overview: rows, columns, data types, missing values\n> - Key distributions and summary statistics\n> - Most interesting correlations\n> - Any outliers or data quality issues\n> - 3-5 insights that jump out\n> \n> Present findings as an interactive HTML report with charts.\"\n\n### Data Cleaning & Transformation\n\nWrangle messy data into shape:\n\n- **Clean Messy Data**: \"Clean this CSV β fix inconsistent date formats, handle missing values, remove duplicates, standardize column names\"\n- **Data Transformation**: \"Pivot this transaction data into a monthly summary by product category\"\n- **Data Merging**: \"Join these three CSV files on customer_id and create a unified dataset\"\n- **Feature Engineering**: \"Create useful features from this raw data for predicting house prices\"\n\n**Example prompt:**\n> \"Clean and transform this dataset:\n> /path/to/messy_data.csv\n> \n> Issues I know about:\n> - Dates are in mixed formats (MM/DD/YYYY and YYYY-MM-DD)\n> - 'Revenue' column has some values with $ signs and commas\n> - Duplicate rows exist\n> - Missing values in 'Region' column\n> \n> Clean it up and give me back a clean CSV plus a summary of what you changed.\"\n\n### Statistical Analysis\n\nRigorous analysis with real numbers:\n\n- **Hypothesis Testing**: \"Is there a statistically significant difference in conversion rates between our A and B variants?\"\n- **Regression Analysis**: \"What factors predict employee salary in this HR dataset? Build a regression model.\"\n- **Time Series Analysis**: \"Analyze this monthly revenue data β trend, seasonality, and forecast next 6 months\"\n- **Cohort Analysis**: \"Create a cohort analysis showing user retention by signup month\"\n\n**Example prompt:**\n> \"I ran an A/B test on our checkout page:\n> /path/to/ab_test_results.csv\n> \n> Columns: user_id, variant (A or B), converted (0/1), revenue, timestamp\n> \n> Tell me:\n> - Is variant B statistically better? (p-value, confidence interval)\n> - Conversion rate difference\n> - Revenue per user difference\n> - Sample size adequacy check\n> - My recommendation: ship B or keep testing?\n> \n> Present with clear charts and a plain-English conclusion.\"\n\n### Visualization & Reporting\n\nTurn data into visual stories:\n\n- **Chart Generation**: \"Create a set of charts showing our quarterly performance from this data\"\n- **Dashboard Reports**: \"Build an interactive dashboard from this sales dataset with filters by region and product\"\n- **Presentation-Ready Visuals**: \"Create publication-quality charts from this research data\"\n- **Comparison Visuals**: \"Visualize how our metrics compare to industry benchmarks\"\n\n### Machine Learning\n\nApplied ML without the setup:\n\n- **Classification**: \"Predict which customers will churn based on this dataset β train a model, show feature importance\"\n- **Clustering**: \"Segment these customers into groups based on behavior β how many natural clusters exist?\"\n- **Forecasting**: \"Forecast next quarter's sales using this historical data\"\n- **Model Evaluation**: \"I trained a model β here are the predictions. Evaluate: accuracy, precision, recall, confusion matrix, ROC curve\"\n\n**Example prompt:**\n> \"Predict customer churn from this dataset:\n> /path/to/customer_features.csv\n> \n> Target column: 'churned'\n> \n> - Train a model, try at least 2 algorithms\n> - Show feature importance β what drives churn?\n> - Confusion matrix and ROC curve\n> - Plain-English summary: 'The top 3 reasons customers churn are...'\n> - Actionable recommendations based on findings\n> \n> I want insights, not just metrics.\"\n\n---\n\n## Supported Data Formats\n\n| Format | How to Send |\n|--------|-------------|\n| **CSV** | Upload via SHOW_FILE |\n| **Excel (XLSX)** | Upload via SHOW_FILE |\n| **JSON** | Upload via SHOW_FILE |\n| **Parquet** | Upload via SHOW_FILE |\n| **SQL exports** | Upload the dump via SHOW_FILE |\n| **Inline data** | Describe small datasets directly in prompt |\n\n---\n\n## Output Formats\n\n| Format | Best For |\n|--------|----------|\n| **Interactive HTML Dashboard** | Explorable charts, filters, drill-downs |\n| **PDF Report** | Shareable analysis reports with charts and findings |\n| **Clean CSV/XLSX** | Cleaned or transformed data files for downstream use |\n| **Markdown** | Quick insights for integration into docs |\n\n---\n\n## Chat Mode for Data\n\n| Scenario | Recommended Mode |\n|----------|------------------|\n| Quick data cleaning, simple charts, basic statistics | `\"agent\"` |\n| Deep analysis with multiple techniques, ML modeling, comprehensive reports | `\"agent team\"` |\n\n**Use `\"agent\"` for most data work.** Data cleaning, EDA, chart generation, and standard statistical analysis execute well in agent mode.\n\n**Use `\"agent team\"` for complex analytical projects** β multi-technique analysis, ML model comparisons, or when you need deep domain reasoning about what the data means.\n\n---\n\n## Example Prompts\n\n**Minimal prompt, maximum insight:**\n> \"Analyze this:\n> /path/to/data.csv\n> \n> Tell me everything interesting.\"\n\nThat's it. CellCog's coding agent will profile the data, run exploratory analysis, find patterns, and present findings with charts. You don't need to know what to ask β the agent figures it out.\n\n**Business analysis:**\n> \"Analyze our e-commerce data:\n> /path/to/orders.csv\n> \n> I need:\n> - Revenue trends (daily, weekly, monthly)\n> - Best and worst performing products\n> - Customer purchase frequency distribution\n> - Average order value trends\n> - Seasonal patterns\n> - Top 5 actionable insights for growing revenue\n> \n> Interactive HTML dashboard with all charts.\"\n\n**Research data analysis:**\n> \"Analyze this survey data from 500 respondents:\n> /path/to/survey.csv\n> \n> Research questions:\n> 1. Is there a significant relationship between age group and product preference?\n> 2. Do satisfaction scores differ by region? (ANOVA)\n> 3. What factors best predict likelihood to recommend? (regression)\n> \n> Include: statistical tests, p-values, effect sizes, and publication-ready charts.\n> PDF report format.\"\n\n---\n\n## Tips for Better Data Analysis\n\n1. **Just upload and ask**: You don't need to describe every column. CellCog reads the data and figures out what's there.\n\n2. **State your question**: \"What drives churn?\" is more focused than \"Analyze this data.\" Both work, but the first gets faster results.\n\n3. **Mention the audience**: \"For my CEO\" means executive summary. \"For the data team\" means show the methodology.\n\n4. **Specify what you'll do with it**: \"I need to present this to the board\" vs \"I need clean data for my ML pipeline\" β context shapes the output.\n\n5. **Don't over-specify methods**: Let CellCog choose the right statistical approach. Say what you want to *learn*, not which algorithm to use.\n\n6. **Iterate**: Upload data β get initial analysis β ask follow-up questions β go deeper. CellCog maintains context across messages.\n\n---\n\n## If CellCog is not installed\n\nRun `/cellcog-setup` (or `/cellcog:cellcog-setup` depending on your tool) to install and authenticate.\n**OpenClaw users:** Run `clawhub install cellcog` instead.\n**Manual setup:** `pip install -U cellcog` and set `CELLCOG_API_KEY`. See the **cellcog** skill for SDK reference.\n\nFile v1.0.13:_meta.json\n\n{\n \"ownerId\": \"kn7a96cj9q65e0bhmzahv790en80ffqm\",\n \"slug\": \"data-analysis-cellcog\",\n \"version\": \"1.0.13\",\n \"publishedAt\": 1785703595540\n}\n\nFile v1.0.13:skill-card.md\n\n## Description:
\nAI data analysis and visualization powered by CellCog for uploaded datasets, including cleaning, exploratory analysis, statistical reporting, model evaluation, charts, and dashboards.
\n\nThis skill is ready for commercial/non-commercial use.
\n\n## Publisher:
\n[cellcog](https://clawhub.ai/user/cellcog)
\n\n### License/Terms of Use:
\nMIT-0
\n\n\n## Use Case:
\nDevelopers, analysts, and data teams use this skill to send permitted datasets to CellCog for automated data cleaning, exploratory analysis, visualization, statistical reporting, and model evaluation.
\n\n### Deployment Geography for Use:
\nGlobal
\n\n## Known Risks and Mitigations:
\nRisk: Selected datasets may be sent to CellCog for automated analysis.
\nMitigation: Use only datasets your organization permits, and do not upload secrets, regulated personal data, proprietary customer data, or untrusted files unless CellCog privacy, retention, and execution controls are understood and approved.
\n\n\n## Reference(s):
\n- [ClawHub skill page](https://clawhub.ai/cellcog/skills/data-analysis-cellcog)
\n- [CellCog homepage](https://cellcog.ai)
\n\n\n## Skill Output:
\n**Output Type(s):** [Text, Markdown, Code, Shell commands, Configuration, Guidance, Files]
\n**Output Format:** [Markdown guidance with Python examples and generated analysis artifacts such as reports, dashboards, charts, and cleaned data files.]
\n**Output Parameters:** [1D]
\n**Other Properties Related to Output:** [May require python3, the cellcog package, and CELLCOG_API_KEY; uploaded datasets are processed by CellCog.]
\n\n## Skill Version(s):
\n1.0.13 (source: server release evidence)
\n\n## Ethical Considerations:
\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.
\n\nArchive v1.0.12: 3 files, 5988 bytes\n\nFiles: skill-card.md (2323b), SKILL.md (10274b), _meta.json (141b)\n\nFile v1.0.12:SKILL.md\n\n---\nname: data-analysis-cellcog\ndescription: \"AI data analysis and visualization powered by CellCog. Data cleaning, exploratory analysis, hypothesis testing, statistical reports, ML model evaluation, dataset profiling, charts, dashboards. Upload CSVs and get analysis back with full Python access.\"\nmetadata:\n openclaw:\n emoji: \"π’\"\n os: [darwin, linux, windows]\n requires:\n bins: [python3]\n env: [CELLCOG_API_KEY]\nauthor: CellCog\nhomepage: https://cellcog.ai\ndependencies: [cellcog]\n---\n# Data Analysis - Your Data Has Answers, CellCog Finds Them\n\nData analysis and visualization from uploaded files.\n\nMost AI tools return code when you ask about data. CellCog returns answers β actual charts, clean datasets, statistical reports, and visual dashboards. Upload messy CSVs with a minimal prompt, and CellCog's coding agent explores your data, finds the patterns, and presents them beautifully. Full Python access for everything from data cleaning to ML model evaluation.\n\n## How to Use\n\nFor your first CellCog task in a session, read the **cellcog** skill for the full SDK reference β file handling, chat modes, timeouts, and more.\n\n**OpenClaw (fire-and-forget):**\n```python\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n notify_session_key=\"agent:main:main\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n)\n```\n\n**All agents except OpenClaw (blocks until done):**\n```python\nfrom cellcog import CellCogClient\nclient = CellCogClient(agent_provider=\"openclaw|cursor|claude-code|codex|...\")\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n)\nprint(result[\"message\"])\n```\n\n\n---\n\n## What Makes Data-Cog Different\n\n### Code as Tool, Not as Output\n\nOther AI tools give you Python code and say \"run this.\" CellCog **runs the code for you** and delivers the results:\n\n| Other AI Tools | Data-Cog |\n|---------------|----------|\n| \"Here's a pandas script to analyze your data\" | Here are your actual insights with charts |\n| \"Run this matplotlib code to see the chart\" | Here's the chart, annotated with findings |\n| \"This SQL query will find outliers\" | Found 23 outliers, here's what they mean |\n| \"You'll need scikit-learn for this\" | Model trained, here's accuracy and feature importance |\n\nYou upload data. You get answers. The code runs behind the scenes.\n\n---\n\n## What Data Work You Can Do\n\n### Exploratory Data Analysis\n\nUnderstand your data fast:\n\n- **Dataset Profiling**: \"Analyze this CSV β distributions, missing values, outliers, correlations, and data quality summary\"\n- **Pattern Discovery**: \"What patterns and trends exist in this sales data? Surprise me.\"\n- **Anomaly Detection**: \"Find unusual patterns in this server log data β what looks abnormal?\"\n- **Relationship Analysis**: \"What factors most strongly correlate with customer churn in this dataset?\"\n\n**Example prompt:**\n> \"Analyze this dataset:\n> /path/to/customer_data.csv\n> \n> I don't know much about this data yet. Give me:\n> - Overview: rows, columns, data types, missing values\n> - Key distributions and summary statistics\n> - Most interesting correlations\n> - Any outliers or data quality issues\n> - 3-5 insights that jump out\n> \n> Present findings as an interactive HTML report with charts.\"\n\n### Data Cleaning & Transformation\n\nWrangle messy data into shape:\n\n- **Clean Messy Data**: \"Clean this CSV β fix inconsistent date formats, handle missing values, remove duplicates, standardize column names\"\n- **Data Transformation**: \"Pivot this transaction data into a monthly summary by product category\"\n- **Data Merging**: \"Join these three CSV files on customer_id and create a unified dataset\"\n- **Feature Engineering**: \"Create useful features from this raw data for predicting house prices\"\n\n**Example prompt:**\n> \"Clean and transform this dataset:\n> /path/to/messy_data.csv\n> \n> Issues I know about:\n> - Dates are in mixed formats (MM/DD/YYYY and YYYY-MM-DD)\n> - 'Revenue' column has some values with $ signs and commas\n> - Duplicate rows exist\n> - Missing values in 'Region' column\n> \n> Clean it up and give me back a clean CSV plus a summary of what you changed.\"\n\n### Statistical Analysis\n\nRigorous analysis with real numbers:\n\n- **Hypothesis Testing**: \"Is there a statistically significant difference in conversion rates between our A and B variants?\"\n- **Regression Analysis**: \"What factors predict employee salary in this HR dataset? Build a regression model.\"\n- **Time Series Analysis**: \"Analyze this monthly revenue data β trend, seasonality, and forecast next 6 months\"\n- **Cohort Analysis**: \"Create a cohort analysis showing user retention by signup month\"\n\n**Example prompt:**\n> \"I ran an A/B test on our checkout page:\n> /path/to/ab_test_results.csv\n> \n> Columns: user_id, variant (A or B), converted (0/1), revenue, timestamp\n> \n> Tell me:\n> - Is variant B statistically better? (p-value, confidence interval)\n> - Conversion rate difference\n> - Revenue per user difference\n> - Sample size adequacy check\n> - My recommendation: ship B or keep testing?\n> \n> Present with clear charts and a plain-English conclusion.\"\n\n### Visualization & Reporting\n\nTurn data into visual stories:\n\n- **Chart Generation**: \"Create a set of charts showing our quarterly performance from this data\"\n- **Dashboard Reports**: \"Build an interactive dashboard from this sales dataset with filters by region and product\"\n- **Presentation-Ready Visuals**: \"Create publication-quality charts from this research data\"\n- **Comparison Visuals**: \"Visualize how our metrics compare to industry benchmarks\"\n\n### Machine Learning\n\nApplied ML without the setup:\n\n- **Classification**: \"Predict which customers will churn based on this dataset β train a model, show feature importance\"\n- **Clustering**: \"Segment these customers into groups based on behavior β how many natural clusters exist?\"\n- **Forecasting**: \"Forecast next quarter's sales using this historical data\"\n- **Model Evaluation**: \"I trained a model β here are the predictions. Evaluate: accuracy, precision, recall, confusion matrix, ROC curve\"\n\n**Example prompt:**\n> \"Predict customer churn from this dataset:\n> /path/to/customer_features.csv\n> \n> Target column: 'churned'\n> \n> - Train a model, try at least 2 algorithms\n> - Show feature importance β what drives churn?\n> - Confusion matrix and ROC curve\n> - Plain-English summary: 'The top 3 reasons customers churn are...'\n> - Actionable recommendations based on findings\n> \n> I want insights, not just metrics.\"\n\n---\n\n## Supported Data Formats\n\n| Format | How to Send |\n|--------|-------------|\n| **CSV** | Upload via SHOW_FILE |\n| **Excel (XLSX)** | Upload via SHOW_FILE |\n| **JSON** | Upload via SHOW_FILE |\n| **Parquet** | Upload via SHOW_FILE |\n| **SQL exports** | Upload the dump via SHOW_FILE |\n| **Inline data** | Describe small datasets directly in prompt |\n\n---\n\n## Output Formats\n\n| Format | Best For |\n|--------|----------|\n| **Interactive HTML Dashboard** | Explorable charts, filters, drill-downs |\n| **PDF Report** | Shareable analysis reports with charts and findings |\n| **Clean CSV/XLSX** | Cleaned or transformed data files for downstream use |\n| **Markdown** | Quick insights for integration into docs |\n\n---\n\n## Chat Mode for Data\n\n| Scenario | Recommended Mode |\n|----------|------------------|\n| Quick data cleaning, simple charts, basic statistics | `\"agent\"` |\n| Deep analysis with multiple techniques, ML modeling, comprehensive reports | `\"agent team\"` |\n\n**Use `\"agent\"` for most data work.** Data cleaning, EDA, chart generation, and standard statistical analysis execute well in agent mode.\n\n**Use `\"agent team\"` for complex analytical projects** β multi-technique analysis, ML model comparisons, or when you need deep domain reasoning about what the data means.\n\n---\n\n## Example Prompts\n\n**Minimal prompt, maximum insight:**\n> \"Analyze this:\n> /path/to/data.csv\n> \n> Tell me everything interesting.\"\n\nThat's it. CellCog's coding agent will profile the data, run exploratory analysis, find patterns, and present findings with charts. You don't need to know what to ask β the agent figures it out.\n\n**Business analysis:**\n> \"Analyze our e-commerce data:\n> /path/to/orders.csv\n> \n> I need:\n> - Revenue trends (daily, weekly, monthly)\n> - Best and worst performing products\n> - Customer purchase frequency distribution\n> - Average order value trends\n> - Seasonal patterns\n> - Top 5 actionable insights for growing revenue\n> \n> Interactive HTML dashboard with all charts.\"\n\n**Research data analysis:**\n> \"Analyze this survey data from 500 respondents:\n> /path/to/survey.csv\n> \n> Research questions:\n> 1. Is there a significant relationship between age group and product preference?\n> 2. Do satisfaction scores differ by region? (ANOVA)\n> 3. What factors best predict likelihood to recommend? (regression)\n> \n> Include: statistical tests, p-values, effect sizes, and publication-ready charts.\n> PDF report format.\"\n\n---\n\n## Tips for Better Data Analysis\n\n1. **Just upload and ask**: You don't need to describe every column. CellCog reads the data and figures out what's there.\n\n2. **State your question**: \"What drives churn?\" is more focused than \"Analyze this data.\" Both work, but the first gets faster results.\n\n3. **Mention the audience**: \"For my CEO\" means executive summary. \"For the data team\" means show the methodology.\n\n4. **Specify what you'll do with it**: \"I need to present this to the board\" vs \"I need clean data for my ML pipeline\" β context shapes the output.\n\n5. **Don't over-specify methods**: Let CellCog choose the right statistical approach. Say what you want to *learn*, not which algorithm to use.\n\n6. **Iterate**: Upload data β get initial analysis β ask follow-up questions β go deeper. CellCog maintains context across messages.\n\n---\n\n## If CellCog is not installed\n\nRun `/cellcog-setup` (or `/cellcog:cellcog-setup` depending on your tool) to install and authenticate.\n**OpenClaw users:** Run `clawhub install cellcog` instead.\n**Manual setup:** `pip install -U cellcog` and set `CELLCOG_API_KEY`. See the **cellcog** skill for SDK reference.\n\nFile v1.0.12:_meta.json\n\n{\n \"ownerId\": \"kn7a96cj9q65e0bhmzahv790en80ffqm\",\n \"slug\": \"data-analysis-cellcog\",\n \"version\": \"1.0.12\",\n \"publishedAt\": 1784259645480\n}\n\nFile v1.0.12:skill-card.md\n\n## Description:
\nAI data analysis and visualization powered by CellCog. Data cleaning, exploratory analysis, hypothesis testing, statistical reports, ML model evaluation, dataset profiling, charts, dashboards. Upload CSVs and get analysis back with full Python access.
\n\nThis skill is ready for commercial/non-commercial use.
\n\n## Publisher:
\n[nitishgargiitd](https://clawhub.ai/user/nitishgargiitd)
\n\n### License/Terms of Use:
\nMIT-0
\n\n\n## Use Case:
\nDevelopers, analysts, and business users use this skill to send uploaded datasets to CellCog for cleaning, exploratory analysis, hypothesis testing, statistical reporting, visualization, dashboard generation, and ML model evaluation.
\n\n### Deployment Geography for Use:
\nGlobal
\n\n## Known Risks and Mitigations:
\nRisk: Uploaded datasets and prompts are processed by CellCog.
\nMitigation: Use the skill only when CellCog is approved for the dataset, prompt content, and intended use.
\nRisk: Users may accidentally upload secrets, regulated personal data, or confidential business data.
\nMitigation: Remove sensitive data before upload unless organizational approval covers that data category.
\nRisk: Cleaned or transformed datasets may change source data.
\nMitigation: Keep original datasets and review transformed outputs before downstream use.
\n\n\n## Reference(s):
\n- [CellCog homepage](https://cellcog.ai)
\n- [ClawHub skill page](https://clawhub.ai/nitishgargiitd/skills/data-analysis-cellcog)
\n\n\n## Skill Output:
\n**Output Type(s):** [text, markdown, code, shell commands, configuration, files]
\n**Output Format:** [Markdown guidance with Python code examples; CellCog results may include HTML dashboards, PDF reports, cleaned CSV/XLSX files, and Markdown summaries.]
\n**Output Parameters:** [1D]
\n**Other Properties Related to Output:** [Requires python3 and CELLCOG_API_KEY; supports darwin, linux, and windows.]
\n\n## Skill Version(s):
\n1.0.12 (source: server release metadata)
\n\n## Ethical Considerations:
\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.
\n\nArchive v1.0.11: 3 files, 6010 bytes\n\nFiles: skill-card.md (2340b), SKILL.md (10256b), _meta.json (141b)\n\nFile v1.0.11:SKILL.md\n\n---\nname: data-cog\ndescription: \"AI data analysis and visualization powered by CellCog. Data cleaning, exploratory analysis, hypothesis testing, statistical reports, ML model evaluation, dataset profiling, charts, dashboards. Upload CSVs and get analysis back with full Python access.\"\nmetadata:\n openclaw:\n emoji: \"π’\"\n os: [darwin, linux, windows]\n requires:\n bins: [python3]\n env: [CELLCOG_API_KEY]\nauthor: CellCog\nhomepage: https://cellcog.ai\ndependencies: [cellcog]\n---\n# Data Cog - Your Data Has Answers, CellCog Finds Them\n\nData analysis and visualization from uploaded files.\n\nMost AI tools return code when you ask about data. CellCog returns answers β actual charts, clean datasets, statistical reports, and visual dashboards. Upload messy CSVs with a minimal prompt, and CellCog's coding agent explores your data, finds the patterns, and presents them beautifully. Full Python access for everything from data cleaning to ML model evaluation.\n\n## How to Use\n\nFor your first CellCog task in a session, read the **cellcog** skill for the full SDK reference β file handling, chat modes, timeouts, and more.\n\n**OpenClaw (fire-and-forget):**\n```python\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n notify_session_key=\"agent:main:main\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n)\n```\n\n**All agents except OpenClaw (blocks until done):**\n```python\nfrom cellcog import CellCogClient\nclient = CellCogClient(agent_provider=\"openclaw|cursor|claude-code|codex|...\")\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n)\nprint(result[\"message\"])\n```\n\n\n---\n\n## What Makes Data-Cog Different\n\n### Code as Tool, Not as Output\n\nOther AI tools give you Python code and say \"run this.\" CellCog **runs the code for you** and delivers the results:\n\n| Other AI Tools | Data-Cog |\n|---------------|----------|\n| \"Here's a pandas script to analyze your data\" | Here are your actual insights with charts |\n| \"Run this matplotlib code to see the chart\" | Here's the chart, annotated with findings |\n| \"This SQL query will find outliers\" | Found 23 outliers, here's what they mean |\n| \"You'll need scikit-learn for this\" | Model trained, here's accuracy and feature importance |\n\nYou upload data. You get answers. The code runs behind the scenes.\n\n---\n\n## What Data Work You Can Do\n\n### Exploratory Data Analysis\n\nUnderstand your data fast:\n\n- **Dataset Profiling**: \"Analyze this CSV β distributions, missing values, outliers, correlations, and data quality summary\"\n- **Pattern Discovery**: \"What patterns and trends exist in this sales data? Surprise me.\"\n- **Anomaly Detection**: \"Find unusual patterns in this server log data β what looks abnormal?\"\n- **Relationship Analysis**: \"What factors most strongly correlate with customer churn in this dataset?\"\n\n**Example prompt:**\n> \"Analyze this dataset:\n> /path/to/customer_data.csv\n> \n> I don't know much about this data yet. Give me:\n> - Overview: rows, columns, data types, missing values\n> - Key distributions and summary statistics\n> - Most interesting correlations\n> - Any outliers or data quality issues\n> - 3-5 insights that jump out\n> \n> Present findings as an interactive HTML report with charts.\"\n\n### Data Cleaning & Transformation\n\nWrangle messy data into shape:\n\n- **Clean Messy Data**: \"Clean this CSV β fix inconsistent date formats, handle missing values, remove duplicates, standardize column names\"\n- **Data Transformation**: \"Pivot this transaction data into a monthly summary by product category\"\n- **Data Merging**: \"Join these three CSV files on customer_id and create a unified dataset\"\n- **Feature Engineering**: \"Create useful features from this raw data for predicting house prices\"\n\n**Example prompt:**\n> \"Clean and transform this dataset:\n> /path/to/messy_data.csv\n> \n> Issues I know about:\n> - Dates are in mixed formats (MM/DD/YYYY and YYYY-MM-DD)\n> - 'Revenue' column has some values with $ signs and commas\n> - Duplicate rows exist\n> - Missing values in 'Region' column\n> \n> Clean it up and give me back a clean CSV plus a summary of what you changed.\"\n\n### Statistical Analysis\n\nRigorous analysis with real numbers:\n\n- **Hypothesis Testing**: \"Is there a statistically significant difference in conversion rates between our A and B variants?\"\n- **Regression Analysis**: \"What factors predict employee salary in this HR dataset? Build a regression model.\"\n- **Time Series Analysis**: \"Analyze this monthly revenue data β trend, seasonality, and forecast next 6 months\"\n- **Cohort Analysis**: \"Create a cohort analysis showing user retention by signup month\"\n\n**Example prompt:**\n> \"I ran an A/B test on our checkout page:\n> /path/to/ab_test_results.csv\n> \n> Columns: user_id, variant (A or B), converted (0/1), revenue, timestamp\n> \n> Tell me:\n> - Is variant B statistically better? (p-value, confidence interval)\n> - Conversion rate difference\n> - Revenue per user difference\n> - Sample size adequacy check\n> - My recommendation: ship B or keep testing?\n> \n> Present with clear charts and a plain-English conclusion.\"\n\n### Visualization & Reporting\n\nTurn data into visual stories:\n\n- **Chart Generation**: \"Create a set of charts showing our quarterly performance from this data\"\n- **Dashboard Reports**: \"Build an interactive dashboard from this sales dataset with filters by region and product\"\n- **Presentation-Ready Visuals**: \"Create publication-quality charts from this research data\"\n- **Comparison Visuals**: \"Visualize how our metrics compare to industry benchmarks\"\n\n### Machine Learning\n\nApplied ML without the setup:\n\n- **Classification**: \"Predict which customers will churn based on this dataset β train a model, show feature importance\"\n- **Clustering**: \"Segment these customers into groups based on behavior β how many natural clusters exist?\"\n- **Forecasting**: \"Forecast next quarter's sales using this historical data\"\n- **Model Evaluation**: \"I trained a model β here are the predictions. Evaluate: accuracy, precision, recall, confusion matrix, ROC curve\"\n\n**Example prompt:**\n> \"Predict customer churn from this dataset:\n> /path/to/customer_features.csv\n> \n> Target column: 'churned'\n> \n> - Train a model, try at least 2 algorithms\n> - Show feature importance β what drives churn?\n> - Confusion matrix and ROC curve\n> - Plain-English summary: 'The top 3 reasons customers churn are...'\n> - Actionable recommendations based on findings\n> \n> I want insights, not just metrics.\"\n\n---\n\n## Supported Data Formats\n\n| Format | How to Send |\n|--------|-------------|\n| **CSV** | Upload via SHOW_FILE |\n| **Excel (XLSX)** | Upload via SHOW_FILE |\n| **JSON** | Upload via SHOW_FILE |\n| **Parquet** | Upload via SHOW_FILE |\n| **SQL exports** | Upload the dump via SHOW_FILE |\n| **Inline data** | Describe small datasets directly in prompt |\n\n---\n\n## Output Formats\n\n| Format | Best For |\n|--------|----------|\n| **Interactive HTML Dashboard** | Explorable charts, filters, drill-downs |\n| **PDF Report** | Shareable analysis reports with charts and findings |\n| **Clean CSV/XLSX** | Cleaned or transformed data files for downstream use |\n| **Markdown** | Quick insights for integration into docs |\n\n---\n\n## Chat Mode for Data\n\n| Scenario | Recommended Mode |\n|----------|------------------|\n| Quick data cleaning, simple charts, basic statistics | `\"agent\"` |\n| Deep analysis with multiple techniques, ML modeling, comprehensive reports | `\"agent team\"` |\n\n**Use `\"agent\"` for most data work.** Data cleaning, EDA, chart generation, and standard statistical analysis execute well in agent mode.\n\n**Use `\"agent team\"` for complex analytical projects** β multi-technique analysis, ML model comparisons, or when you need deep domain reasoning about what the data means.\n\n---\n\n## Example Prompts\n\n**Minimal prompt, maximum insight:**\n> \"Analyze this:\n> /path/to/data.csv\n> \n> Tell me everything interesting.\"\n\nThat's it. CellCog's coding agent will profile the data, run exploratory analysis, find patterns, and present findings with charts. You don't need to know what to ask β the agent figures it out.\n\n**Business analysis:**\n> \"Analyze our e-commerce data:\n> /path/to/orders.csv\n> \n> I need:\n> - Revenue trends (daily, weekly, monthly)\n> - Best and worst performing products\n> - Customer purchase frequency distribution\n> - Average order value trends\n> - Seasonal patterns\n> - Top 5 actionable insights for growing revenue\n> \n> Interactive HTML dashboard with all charts.\"\n\n**Research data analysis:**\n> \"Analyze this survey data from 500 respondents:\n> /path/to/survey.csv\n> \n> Research questions:\n> 1. Is there a significant relationship between age group and product preference?\n> 2. Do satisfaction scores differ by region? (ANOVA)\n> 3. What factors best predict likelihood to recommend? (regression)\n> \n> Include: statistical tests, p-values, effect sizes, and publication-ready charts.\n> PDF report format.\"\n\n---\n\n## Tips for Better Data Analysis\n\n1. **Just upload and ask**: You don't need to describe every column. CellCog reads the data and figures out what's there.\n\n2. **State your question**: \"What drives churn?\" is more focused than \"Analyze this data.\" Both work, but the first gets faster results.\n\n3. **Mention the audience**: \"For my CEO\" means executive summary. \"For the data team\" means show the methodology.\n\n4. **Specify what you'll do with it**: \"I need to present this to the board\" vs \"I need clean data for my ML pipeline\" β context shapes the output.\n\n5. **Don't over-specify methods**: Let CellCog choose the right statistical approach. Say what you want to *learn*, not which algorithm to use.\n\n6. **Iterate**: Upload data β get initial analysis β ask follow-up questions β go deeper. CellCog maintains context across messages.\n\n---\n\n## If CellCog is not installed\n\nRun `/cellcog-setup` (or `/cellcog:cellcog-setup` depending on your tool) to install and authenticate.\n**OpenClaw users:** Run `clawhub install cellcog` instead.\n**Manual setup:** `pip install -U cellcog` and set `CELLCOG_API_KEY`. See the **cellcog** skill for SDK reference.\n\nFile v1.0.11:_meta.json\n\n{\n \"ownerId\": \"kn7a96cj9q65e0bhmzahv790en80ffqm\",\n \"slug\": \"data-analysis-cellcog\",\n \"version\": \"1.0.11\",\n \"publishedAt\": 1776925128710\n}\n\nFile v1.0.11:skill-card.md\n\n## Description:
\nAI data analysis and visualization powered by CellCog, including data cleaning, exploratory analysis, hypothesis testing, statistical reports, ML model evaluation, dataset profiling, charts, and dashboards.
\n\nThis skill is ready for commercial/non-commercial use.
\n\n## Publisher:
\n[nitishgargiitd](https://clawhub.ai/user/nitishgargiitd)
\n\n### License/Terms of Use:
\nMIT-0
\n\n\n## Use Case:
\nDevelopers, analysts, and data teams use Data Cog to send datasets to CellCog for data cleaning, exploratory analysis, statistical testing, visualization, reporting, and machine learning evaluation.
\n\n### Deployment Geography for Use:
\nGlobal
\n\n## Known Risks and Mitigations:
\nRisk: Uploaded datasets may be shared with an external CellCog analysis service.
\nMitigation: Do not upload secrets, credentials, personal data, health or financial records, or confidential business datasets unless CellCog's privacy, retention, and data-processing terms have been approved.
\nRisk: Generated analyses, statistical conclusions, charts, or model evaluations may be incorrect or misleading if the data, assumptions, or prompt are incomplete.
\nMitigation: Review outputs, validate methods and assumptions, and confirm important findings before using them for business, scientific, legal, financial, or operational decisions.
\n\n\n## Reference(s):
\n- [CellCog](https://cellcog.ai)
\n- [Data Cog on ClawHub](https://clawhub.ai/nitishgargiitd/skills/data-cog)
\n\n\n## Skill Output:
\n**Output Type(s):** [Analysis, Markdown, Code, Files, Configuration instructions, Guidance]
\n**Output Format:** [Markdown with analysis summaries, charts, reports, clean data files, dashboards, and Python-backed results]
\n**Output Parameters:** [1D]
\n**Other Properties Related to Output:** [Requires python3, the cellcog dependency, and CELLCOG_API_KEY; outputs may be generated by an external CellCog analysis service.]
\n\n## Skill Version(s):
\n1.0.11 (source: server release evidence)
\n\n## Ethical Considerations:
\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.
\n\nArchive v1.0.10: 2 files, 4751 bytes\n\nFiles: SKILL.md (10191b), _meta.json (141b)\n\nFile v1.0.10:SKILL.md\n\n---\nname: data-cog\ndescription: \"AI data analysis and visualization powered by CellCog. Data cleaning, exploratory analysis, hypothesis testing, statistical reports, ML model evaluation, dataset profiling, charts, dashboards. Upload CSVs and get analysis back with full Python access.\"\nmetadata:\n openclaw:\n emoji: \"π’\"\n os: [darwin, linux, windows]\nauthor: CellCog\nhomepage: https://cellcog.ai\ndependencies: [cellcog]\n---\n# Data Cog - Your Data Has Answers, CellCog Finds Them\n\nData analysis and visualization from uploaded files.\n\nMost AI tools return code when you ask about data. CellCog returns answers β actual charts, clean datasets, statistical reports, and visual dashboards. Upload messy CSVs with a minimal prompt, and CellCog's coding agent explores your data, finds the patterns, and presents them beautifully. Full Python access for everything from data cleaning to ML model evaluation.\n\n## How to Use\n\nFor your first CellCog task in a session, read the **cellcog** skill for the full SDK reference β file handling, chat modes, timeouts, and more.\n\n**OpenClaw (fire-and-forget):**\n```python\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n notify_session_key=\"agent:main:main\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n)\n```\n\n**All agents except OpenClaw (blocks until done):**\n```python\nfrom cellcog import CellCogClient\nclient = CellCogClient(agent_provider=\"openclaw|cursor|claude-code|codex|...\")\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n)\nprint(result[\"message\"])\n```\n\n\n---\n\n## What Makes Data-Cog Different\n\n### Code as Tool, Not as Output\n\nOther AI tools give you Python code and say \"run this.\" CellCog **runs the code for you** and delivers the results:\n\n| Other AI Tools | Data-Cog |\n|---------------|----------|\n| \"Here's a pandas script to analyze your data\" | Here are your actual insights with charts |\n| \"Run this matplotlib code to see the chart\" | Here's the chart, annotated with findings |\n| \"This SQL query will find outliers\" | Found 23 outliers, here's what they mean |\n| \"You'll need scikit-learn for this\" | Model trained, here's accuracy and feature importance |\n\nYou upload data. You get answers. The code runs behind the scenes.\n\n---\n\n## What Data Work You Can Do\n\n### Exploratory Data Analysis\n\nUnderstand your data fast:\n\n- **Dataset Profiling**: \"Analyze this CSV β distributions, missing values, outliers, correlations, and data quality summary\"\n- **Pattern Discovery**: \"What patterns and trends exist in this sales data? Surprise me.\"\n- **Anomaly Detection**: \"Find unusual patterns in this server log data β what looks abnormal?\"\n- **Relationship Analysis**: \"What factors most strongly correlate with customer churn in this dataset?\"\n\n**Example prompt:**\n> \"Analyze this dataset:\n> /path/to/customer_data.csv\n> \n> I don't know much about this data yet. Give me:\n> - Overview: rows, columns, data types, missing values\n> - Key distributions and summary statistics\n> - Most interesting correlations\n> - Any outliers or data quality issues\n> - 3-5 insights that jump out\n> \n> Present findings as an interactive HTML report with charts.\"\n\n### Data Cleaning & Transformation\n\nWrangle messy data into shape:\n\n- **Clean Messy Data**: \"Clean this CSV β fix inconsistent date formats, handle missing values, remove duplicates, standardize column names\"\n- **Data Transformation**: \"Pivot this transaction data into a monthly summary by product category\"\n- **Data Merging**: \"Join these three CSV files on customer_id and create a unified dataset\"\n- **Feature Engineering**: \"Create useful features from this raw data for predicting house prices\"\n\n**Example prompt:**\n> \"Clean and transform this dataset:\n> /path/to/messy_data.csv\n> \n> Issues I know about:\n> - Dates are in mixed formats (MM/DD/YYYY and YYYY-MM-DD)\n> - 'Revenue' column has some values with $ signs and commas\n> - Duplicate rows exist\n> - Missing values in 'Region' column\n> \n> Clean it up and give me back a clean CSV plus a summary of what you changed.\"\n\n### Statistical Analysis\n\nRigorous analysis with real numbers:\n\n- **Hypothesis Testing**: \"Is there a statistically significant difference in conversion rates between our A and B variants?\"\n- **Regression Analysis**: \"What factors predict employee salary in this HR dataset? Build a regression model.\"\n- **Time Series Analysis**: \"Analyze this monthly revenue data β trend, seasonality, and forecast next 6 months\"\n- **Cohort Analysis**: \"Create a cohort analysis showing user retention by signup month\"\n\n**Example prompt:**\n> \"I ran an A/B test on our checkout page:\n> /path/to/ab_test_results.csv\n> \n> Columns: user_id, variant (A or B), converted (0/1), revenue, timestamp\n> \n> Tell me:\n> - Is variant B statistically better? (p-value, confidence interval)\n> - Conversion rate difference\n> - Revenue per user difference\n> - Sample size adequacy check\n> - My recommendation: ship B or keep testing?\n> \n> Present with clear charts and a plain-English conclusion.\"\n\n### Visualization & Reporting\n\nTurn data into visual stories:\n\n- **Chart Generation**: \"Create a set of charts showing our quarterly performance from this data\"\n- **Dashboard Reports**: \"Build an interactive dashboard from this sales dataset with filters by region and product\"\n- **Presentation-Ready Visuals**: \"Create publication-quality charts from this research data\"\n- **Comparison Visuals**: \"Visualize how our metrics compare to industry benchmarks\"\n\n### Machine Learning\n\nApplied ML without the setup:\n\n- **Classification**: \"Predict which customers will churn based on this dataset β train a model, show feature importance\"\n- **Clustering**: \"Segment these customers into groups based on behavior β how many natural clusters exist?\"\n- **Forecasting**: \"Forecast next quarter's sales using this historical data\"\n- **Model Evaluation**: \"I trained a model β here are the predictions. Evaluate: accuracy, precision, recall, confusion matrix, ROC curve\"\n\n**Example prompt:**\n> \"Predict customer churn from this dataset:\n> /path/to/customer_features.csv\n> \n> Target column: 'churned'\n> \n> - Train a model, try at least 2 algorithms\n> - Show feature importance β what drives churn?\n> - Confusion matrix and ROC curve\n> - Plain-English summary: 'The top 3 reasons customers churn are...'\n> - Actionable recommendations based on findings\n> \n> I want insights, not just metrics.\"\n\n---\n\n## Supported Data Formats\n\n| Format | How to Send |\n|--------|-------------|\n| **CSV** | Upload via SHOW_FILE |\n| **Excel (XLSX)** | Upload via SHOW_FILE |\n| **JSON** | Upload via SHOW_FILE |\n| **Parquet** | Upload via SHOW_FILE |\n| **SQL exports** | Upload the dump via SHOW_FILE |\n| **Inline data** | Describe small datasets directly in prompt |\n\n---\n\n## Output Formats\n\n| Format | Best For |\n|--------|----------|\n| **Interactive HTML Dashboard** | Explorable charts, filters, drill-downs |\n| **PDF Report** | Shareable analysis reports with charts and findings |\n| **Clean CSV/XLSX** | Cleaned or transformed data files for downstream use |\n| **Markdown** | Quick insights for integration into docs |\n\n---\n\n## Chat Mode for Data\n\n| Scenario | Recommended Mode |\n|----------|------------------|\n| Quick data cleaning, simple charts, basic statistics | `\"agent\"` |\n| Deep analysis with multiple techniques, ML modeling, comprehensive reports | `\"agent team\"` |\n\n**Use `\"agent\"` for most data work.** Data cleaning, EDA, chart generation, and standard statistical analysis execute well in agent mode.\n\n**Use `\"agent team\"` for complex analytical projects** β multi-technique analysis, ML model comparisons, or when you need deep domain reasoning about what the data means.\n\n---\n\n## Example Prompts\n\n**Minimal prompt, maximum insight:**\n> \"Analyze this:\n> /path/to/data.csv\n> \n> Tell me everything interesting.\"\n\nThat's it. CellCog's coding agent will profile the data, run exploratory analysis, find patterns, and present findings with charts. You don't need to know what to ask β the agent figures it out.\n\n**Business analysis:**\n> \"Analyze our e-commerce data:\n> /path/to/orders.csv\n> \n> I need:\n> - Revenue trends (daily, weekly, monthly)\n> - Best and worst performing products\n> - Customer purchase frequency distribution\n> - Average order value trends\n> - Seasonal patterns\n> - Top 5 actionable insights for growing revenue\n> \n> Interactive HTML dashboard with all charts.\"\n\n**Research data analysis:**\n> \"Analyze this survey data from 500 respondents:\n> /path/to/survey.csv\n> \n> Research questions:\n> 1. Is there a significant relationship between age group and product preference?\n> 2. Do satisfaction scores differ by region? (ANOVA)\n> 3. What factors best predict likelihood to recommend? (regression)\n> \n> Include: statistical tests, p-values, effect sizes, and publication-ready charts.\n> PDF report format.\"\n\n---\n\n## Tips for Better Data Analysis\n\n1. **Just upload and ask**: You don't need to describe every column. CellCog reads the data and figures out what's there.\n\n2. **State your question**: \"What drives churn?\" is more focused than \"Analyze this data.\" Both work, but the first gets faster results.\n\n3. **Mention the audience**: \"For my CEO\" means executive summary. \"For the data team\" means show the methodology.\n\n4. **Specify what you'll do with it**: \"I need to present this to the board\" vs \"I need clean data for my ML pipeline\" β context shapes the output.\n\n5. **Don't over-specify methods**: Let CellCog choose the right statistical approach. Say what you want to *learn*, not which algorithm to use.\n\n6. **Iterate**: Upload data β get initial analysis β ask follow-up questions β go deeper. CellCog maintains context across messages.\n\n---\n\n## If CellCog is not installed\n\nRun `/cellcog-setup` (or `/cellcog:cellcog-setup` depending on your tool) to install and authenticate.\n**OpenClaw users:** Run `clawhub install cellcog` instead.\n**Manual setup:** `pip install -U cellcog` and set `CELLCOG_API_KEY`. See the **cellcog** skill for SDK reference.\n\nFile v1.0.10:_meta.json\n\n{\n \"ownerId\": \"kn7a96cj9q65e0bhmzahv790en80ffqm\",\n \"slug\": \"data-analysis-cellcog\",\n \"version\": \"1.0.10\",\n \"publishedAt\": 1776187927438\n}\n\nArchive v1.0.9: 2 files, 4793 bytes\n\nFiles: SKILL.md (10268b), _meta.json (140b)\n\nFile v1.0.9:SKILL.md\n\n---\nname: data-cog\ndescription: \"AI data analysis and visualization powered by CellCog. Upload CSVs and get charts, dashboards, statistical reports, and clean data back. Data cleaning, exploratory analysis, hypothesis testing, ML model evaluation, dataset profiling, and data visualization. Full Python access. #1 on DeepResearch Bench (Apr 2026). Analyzes everything, presents it beautifully.\"\nmetadata:\n openclaw:\n emoji: \"π’\"\n os: [darwin, linux, windows]\nauthor: CellCog\nhomepage: https://cellcog.ai\ndependencies: [cellcog]\n---\n# Data Cog - Your Data Has Answers, CellCog Finds Them\n\nData analysis and visualization from uploaded files.\n\nMost AI tools return code when you ask about data. CellCog returns answers β actual charts, clean datasets, statistical reports, and visual dashboards. Upload messy CSVs with a minimal prompt, and CellCog's coding agent explores your data, finds the patterns, and presents them beautifully. Full Python access for everything from data cleaning to ML model evaluation.\n\n## How to Use\n\nFor your first CellCog task in a session, read the **cellcog** skill for the full SDK reference β file handling, chat modes, timeouts, and more.\n\n**OpenClaw (fire-and-forget):**\n```python\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n notify_session_key=\"agent:main:main\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n)\n```\n\n**Cursor / Claude Code / Other agents (blocks until done):**\n```python\nfrom cellcog import CellCogClient\nclient = CellCogClient(agent_provider=\"openclaw|cursor|claude-code|codex|...\")\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n)\nprint(result[\"message\"])\n```\n\n\n---\n\n## What Makes Data-Cog Different\n\n### Code as Tool, Not as Output\n\nOther AI tools give you Python code and say \"run this.\" CellCog **runs the code for you** and delivers the results:\n\n| Other AI Tools | Data-Cog |\n|---------------|----------|\n| \"Here's a pandas script to analyze your data\" | Here are your actual insights with charts |\n| \"Run this matplotlib code to see the chart\" | Here's the chart, annotated with findings |\n| \"This SQL query will find outliers\" | Found 23 outliers, here's what they mean |\n| \"You'll need scikit-learn for this\" | Model trained, here's accuracy and feature importance |\n\nYou upload data. You get answers. The code runs behind the scenes.\n\n---\n\n## What Data Work You Can Do\n\n### Exploratory Data Analysis\n\nUnderstand your data fast:\n\n- **Dataset Profiling**: \"Analyze this CSV β distributions, missing values, outliers, correlations, and data quality summary\"\n- **Pattern Discovery**: \"What patterns and trends exist in this sales data? Surprise me.\"\n- **Anomaly Detection**: \"Find unusual patterns in this server log data β what looks abnormal?\"\n- **Relationship Analysis**: \"What factors most strongly correlate with customer churn in this dataset?\"\n\n**Example prompt:**\n> \"Analyze this dataset:\n> /path/to/customer_data.csv\n> \n> I don't know much about this data yet. Give me:\n> - Overview: rows, columns, data types, missing values\n> - Key distributions and summary statistics\n> - Most interesting correlations\n> - Any outliers or data quality issues\n> - 3-5 insights that jump out\n> \n> Present findings as an interactive HTML report with charts.\"\n\n### Data Cleaning & Transformation\n\nWrangle messy data into shape:\n\n- **Clean Messy Data**: \"Clean this CSV β fix inconsistent date formats, handle missing values, remove duplicates, standardize column names\"\n- **Data Transformation**: \"Pivot this transaction data into a monthly summary by product category\"\n- **Data Merging**: \"Join these three CSV files on customer_id and create a unified dataset\"\n- **Feature Engineering**: \"Create useful features from this raw data for predicting house prices\"\n\n**Example prompt:**\n> \"Clean and transform this dataset:\n> /path/to/messy_data.csv\n> \n> Issues I know about:\n> - Dates are in mixed formats (MM/DD/YYYY and YYYY-MM-DD)\n> - 'Revenue' column has some values with $ signs and commas\n> - Duplicate rows exist\n> - Missing values in 'Region' column\n> \n> Clean it up and give me back a clean CSV plus a summary of what you changed.\"\n\n### Statistical Analysis\n\nRigorous analysis with real numbers:\n\n- **Hypothesis Testing**: \"Is there a statistically significant difference in conversion rates between our A and B variants?\"\n- **Regression Analysis**: \"What factors predict employee salary in this HR dataset? Build a regression model.\"\n- **Time Series Analysis**: \"Analyze this monthly revenue data β trend, seasonality, and forecast next 6 months\"\n- **Cohort Analysis**: \"Create a cohort analysis showing user retention by signup month\"\n\n**Example prompt:**\n> \"I ran an A/B test on our checkout page:\n> /path/to/ab_test_results.csv\n> \n> Columns: user_id, variant (A or B), converted (0/1), revenue, timestamp\n> \n> Tell me:\n> - Is variant B statistically better? (p-value, confidence interval)\n> - Conversion rate difference\n> - Revenue per user difference\n> - Sample size adequacy check\n> - My recommendation: ship B or keep testing?\n> \n> Present with clear charts and a plain-English conclusion.\"\n\n### Visualization & Reporting\n\nTurn data into visual stories:\n\n- **Chart Generation**: \"Create a set of charts showing our quarterly performance from this data\"\n- **Dashboard Reports**: \"Build an interactive dashboard from this sales dataset with filters by region and product\"\n- **Presentation-Ready Visuals**: \"Create publication-quality charts from this research data\"\n- **Comparison Visuals**: \"Visualize how our metrics compare to industry benchmarks\"\n\n### Machine Learning\n\nApplied ML without the setup:\n\n- **Classification**: \"Predict which customers will churn based on this dataset β train a model, show feature importance\"\n- **Clustering**: \"Segment these customers into groups based on behavior β how many natural clusters exist?\"\n- **Forecasting**: \"Forecast next quarter's sales using this historical data\"\n- **Model Evaluation**: \"I trained a model β here are the predictions. Evaluate: accuracy, precision, recall, confusion matrix, ROC curve\"\n\n**Example prompt:**\n> \"Predict customer churn from this dataset:\n> /path/to/customer_features.csv\n> \n> Target column: 'churned'\n> \n> - Train a model, try at least 2 algorithms\n> - Show feature importance β what drives churn?\n> - Confusion matrix and ROC curve\n> - Plain-English summary: 'The top 3 reasons customers churn are...'\n> - Actionable recommendations based on findings\n> \n> I want insights, not just metrics.\"\n\n---\n\n## Supported Data Formats\n\n| Format | How to Send |\n|--------|-------------|\n| **CSV** | Upload via SHOW_FILE |\n| **Excel (XLSX)** | Upload via SHOW_FILE |\n| **JSON** | Upload via SHOW_FILE |\n| **Parquet** | Upload via SHOW_FILE |\n| **SQL exports** | Upload the dump via SHOW_FILE |\n| **Inline data** | Describe small datasets directly in prompt |\n\n---\n\n## Output Formats\n\n| Format | Best For |\n|--------|----------|\n| **Interactive HTML Dashboard** | Explorable charts, filters, drill-downs |\n| **PDF Report** | Shareable analysis reports with charts and findings |\n| **Clean CSV/XLSX** | Cleaned or transformed data files for downstream use |\n| **Markdown** | Quick insights for integration into docs |\n\n---\n\n## Chat Mode for Data\n\n| Scenario | Recommended Mode |\n|----------|------------------|\n| Quick data cleaning, simple charts, basic statistics | `\"agent\"` |\n| Deep analysis with multiple techniques, ML modeling, comprehensive reports | `\"agent team\"` |\n\n**Use `\"agent\"` for most data work.** Data cleaning, EDA, chart generation, and standard statistical analysis execute well in agent mode.\n\n**Use `\"agent team\"` for complex analytical projects** β multi-technique analysis, ML model comparisons, or when you need deep domain reasoning about what the data means.\n\n---\n\n## Example Prompts\n\n**Minimal prompt, maximum insight:**\n> \"Analyze this:\n> /path/to/data.csv\n> \n> Tell me everything interesting.\"\n\nThat's it. CellCog's coding agent will profile the data, run exploratory analysis, find patterns, and present findings with charts. You don't need to know what to ask β the agent figures it out.\n\n**Business analysis:**\n> \"Analyze our e-commerce data:\n> /path/to/orders.csv\n> \n> I need:\n> - Revenue trends (daily, weekly, monthly)\n> - Best and worst performing products\n> - Customer purchase frequency distribution\n> - Average order value trends\n> - Seasonal patterns\n> - Top 5 actionable insights for growing revenue\n> \n> Interactive HTML dashboard with all charts.\"\n\n**Research data analysis:**\n> \"Analyze this survey data from 500 respondents:\n> /path/to/survey.csv\n> \n> Research questions:\n> 1. Is there a significant relationship between age group and product preference?\n> 2. Do satisfaction scores differ by region? (ANOVA)\n> 3. What factors best predict likelihood to recommend? (regression)\n> \n> Include: statistical tests, p-values, effect sizes, and publication-ready charts.\n> PDF report format.\"\n\n---\n\n## Tips for Better Data Analysis\n\n1. **Just upload and ask**: You don't need to describe every column. CellCog reads the data and figures out what's there.\n\n2. **State your question**: \"What drives churn?\" is more focused than \"Analyze this data.\" Both work, but the first gets faster results.\n\n3. **Mention the audience**: \"For my CEO\" means executive summary. \"For the data team\" means show the methodology.\n\n4. **Specify what you'll do with it**: \"I need to present this to the board\" vs \"I need clean data for my ML pipeline\" β context shapes the output.\n\n5. **Don't over-specify methods**: Let CellCog choose the right statistical approach. Say what you want to *learn*, not which algorithm to use.\n\n6. **Iterate**: Upload data β get initial analysis β ask follow-up questions β go deeper. CellCog maintains context across messages.\n\n---\n\n## If CellCog is not installed\n\n**Cursor:** Run `/cellcog-setup` to install and authenticate.\n**OpenClaw:** Run `clawhub install cellcog` for SDK setup.\n**Other agents:** `pip install -U cellcog` and set `CELLCOG_API_KEY`. See the **cellcog** skill for SDK reference.\n\nFile v1.0.9:_meta.json\n\n{\n \"ownerId\": \"kn7a96cj9q65e0bhmzahv790en80ffqm\",\n \"slug\": \"data-analysis-cellcog\",\n \"version\": \"1.0.9\",\n \"publishedAt\": 1776042040063\n}\n\nArchive v1.0.8: 2 files, 4667 bytes\n\nFiles: SKILL.md (9992b), _meta.json (140b)\n\nFile v1.0.8:SKILL.md\n\n---\nname: data-cog\ndescription: \"Powered by CellCog. Data analysis and visualization from uploaded files. Upload CSVs for charts, statistical reports, data cleaning, exploratory analysis, hypothesis testing, and dataset profiling.\"\nmetadata:\n openclaw:\n emoji: \"π’\"\n os: [darwin, linux, windows]\nauthor: CellCog\nhomepage: https://cellcog.ai\ndependencies: [cellcog]\n---\n# Data Cog - Your Data Has Answers, CellCog Finds Them\n\nData analysis and visualization from uploaded files.\n\nMost AI tools return code when you ask about data. CellCog returns answers β actual charts, clean datasets, statistical reports, and visual dashboards. Upload messy CSVs with a minimal prompt, and CellCog's coding agent explores your data, finds the patterns, and presents them beautifully. Full Python access for everything from data cleaning to ML model evaluation.\n\n## How to Use\n\nFor your first CellCog task in a session, read the **cellcog** skill for the full SDK reference β file handling, chat modes, timeouts, and more.\n\n**OpenClaw (fire-and-forget):**\n```python\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n notify_session_key=\"agent:main:main\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n)\n```\n\n**Cursor / Claude Code / Other agents (blocks until done):**\n```python\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n)\nprint(result[\"message\"])\n```\n\n\n---\n\n## What Makes Data-Cog Different\n\n### Code as Tool, Not as Output\n\nOther AI tools give you Python code and say \"run this.\" CellCog **runs the code for you** and delivers the results:\n\n| Other AI Tools | Data-Cog |\n|---------------|----------|\n| \"Here's a pandas script to analyze your data\" | Here are your actual insights with charts |\n| \"Run this matplotlib code to see the chart\" | Here's the chart, annotated with findings |\n| \"This SQL query will find outliers\" | Found 23 outliers, here's what they mean |\n| \"You'll need scikit-learn for this\" | Model trained, here's accuracy and feature importance |\n\nYou upload data. You get answers. The code runs behind the scenes.\n\n---\n\n## What Data Work You Can Do\n\n### Exploratory Data Analysis\n\nUnderstand your data fast:\n\n- **Dataset Profiling**: \"Analyze this CSV β distributions, missing values, outliers, correlations, and data quality summary\"\n- **Pattern Discovery**: \"What patterns and trends exist in this sales data? Surprise me.\"\n- **Anomaly Detection**: \"Find unusual patterns in this server log data β what looks abnormal?\"\n- **Relationship Analysis**: \"What factors most strongly correlate with customer churn in this dataset?\"\n\n**Example prompt:**\n> \"Analyze this dataset:\n> /path/to/customer_data.csv\n> \n> I don't know much about this data yet. Give me:\n> - Overview: rows, columns, data types, missing values\n> - Key distributions and summary statistics\n> - Most interesting correlations\n> - Any outliers or data quality issues\n> - 3-5 insights that jump out\n> \n> Present findings as an interactive HTML report with charts.\"\n\n### Data Cleaning & Transformation\n\nWrangle messy data into shape:\n\n- **Clean Messy Data**: \"Clean this CSV β fix inconsistent date formats, handle missing values, remove duplicates, standardize column names\"\n- **Data Transformation**: \"Pivot this transaction data into a monthly summary by product category\"\n- **Data Merging**: \"Join these three CSV files on customer_id and create a unified dataset\"\n- **Feature Engineering**: \"Create useful features from this raw data for predicting house prices\"\n\n**Example prompt:**\n> \"Clean and transform this dataset:\n> /path/to/messy_data.csv\n> \n> Issues I know about:\n> - Dates are in mixed formats (MM/DD/YYYY and YYYY-MM-DD)\n> - 'Revenue' column has some values with $ signs and commas\n> - Duplicate rows exist\n> - Missing values in 'Region' column\n> \n> Clean it up and give me back a clean CSV plus a summary of what you changed.\"\n\n### Statistical Analysis\n\nRigorous analysis with real numbers:\n\n- **Hypothesis Testing**: \"Is there a statistically significant difference in conversion rates between our A and B variants?\"\n- **Regression Analysis**: \"What factors predict employee salary in this HR dataset? Build a regression model.\"\n- **Time Series Analysis**: \"Analyze this monthly revenue data β trend, seasonality, and forecast next 6 months\"\n- **Cohort Analysis**: \"Create a cohort analysis showing user retention by signup month\"\n\n**Example prompt:**\n> \"I ran an A/B test on our checkout page:\n> /path/to/ab_test_results.csv\n> \n> Columns: user_id, variant (A or B), converted (0/1), revenue, timestamp\n> \n> Tell me:\n> - Is variant B statistically better? (p-value, confidence interval)\n> - Conversion rate difference\n> - Revenue per user difference\n> - Sample size adequacy check\n> - My recommendation: ship B or keep testing?\n> \n> Present with clear charts and a plain-English conclusion.\"\n\n### Visualization & Reporting\n\nTurn data into visual stories:\n\n- **Chart Generation**: \"Create a set of charts showing our quarterly performance from this data\"\n- **Dashboard Reports**: \"Build an interactive dashboard from this sales dataset with filters by region and product\"\n- **Presentation-Ready Visuals**: \"Create publication-quality charts from this research data\"\n- **Comparison Visuals**: \"Visualize how our metrics compare to industry benchmarks\"\n\n### Machine Learning\n\nApplied ML without the setup:\n\n- **Classification**: \"Predict which customers will churn based on this dataset β train a model, show feature importance\"\n- **Clustering**: \"Segment these customers into groups based on behavior β how many natural clusters exist?\"\n- **Forecasting**: \"Forecast next quarter's sales using this historical data\"\n- **Model Evaluation**: \"I trained a model β here are the predictions. Evaluate: accuracy, precision, recall, confusion matrix, ROC curve\"\n\n**Example prompt:**\n> \"Predict customer churn from this dataset:\n> /path/to/customer_features.csv\n> \n> Target column: 'churned'\n> \n> - Train a model, try at least 2 algorithms\n> - Show feature importance β what drives churn?\n> - Confusion matrix and ROC curve\n> - Plain-English summary: 'The top 3 reasons customers churn are...'\n> - Actionable recommendations based on findings\n> \n> I want insights, not just metrics.\"\n\n---\n\n## Supported Data Formats\n\n| Format | How to Send |\n|--------|-------------|\n| **CSV** | Upload via SHOW_FILE |\n| **Excel (XLSX)** | Upload via SHOW_FILE |\n| **JSON** | Upload via SHOW_FILE |\n| **Parquet** | Upload via SHOW_FILE |\n| **SQL exports** | Upload the dump via SHOW_FILE |\n| **Inline data** | Describe small datasets directly in prompt |\n\n---\n\n## Output Formats\n\n| Format | Best For |\n|--------|----------|\n| **Interactive HTML Dashboard** | Explorable charts, filters, drill-downs |\n| **PDF Report** | Shareable analysis reports with charts and findings |\n| **Clean CSV/XLSX** | Cleaned or transformed data files for downstream use |\n| **Markdown** | Quick insights for integration into docs |\n\n---\n\n## Chat Mode for Data\n\n| Scenario | Recommended Mode |\n|----------|------------------|\n| Quick data cleaning, simple charts, basic statistics | `\"agent\"` |\n| Deep analysis with multiple techniques, ML modeling, comprehensive reports | `\"agent team\"` |\n\n**Use `\"agent\"` for most data work.** Data cleaning, EDA, chart generation, and standard statistical analysis execute well in agent mode.\n\n**Use `\"agent team\"` for complex analytical projects** β multi-technique analysis, ML model comparisons, or when you need deep domain reasoning about what the data means.\n\n---\n\n## Example Prompts\n\n**Minimal prompt, maximum insight:**\n> \"Analyze this:\n> /path/to/data.csv\n> \n> Tell me everything interesting.\"\n\nThat's it. CellCog's coding agent will profile the data, run exploratory analysis, find patterns, and present findings with charts. You don't need to know what to ask β the agent figures it out.\n\n**Business analysis:**\n> \"Analyze our e-commerce data:\n> /path/to/orders.csv\n> \n> I need:\n> - Revenue trends (daily, weekly, monthly)\n> - Best and worst performing products\n> - Customer purchase frequency distribution\n> - Average order value trends\n> - Seasonal patterns\n> - Top 5 actionable insights for growing revenue\n> \n> Interactive HTML dashboard with all charts.\"\n\n**Research data analysis:**\n> \"Analyze this survey data from 500 respondents:\n> /path/to/survey.csv\n> \n> Research questions:\n> 1. Is there a significant relationship between age group and product preference?\n> 2. Do satisfaction scores differ by region? (ANOVA)\n> 3. What factors best predict likelihood to recommend? (regression)\n> \n> Include: statistical tests, p-values, effect sizes, and publication-ready charts.\n> PDF report format.\"\n\n---\n\n## Tips for Better Data Analysis\n\n1. **Just upload and ask**: You don't need to describe every column. CellCog reads the data and figures out what's there.\n\n2. **State your question**: \"What drives churn?\" is more focused than \"Analyze this data.\" Both work, but the first gets faster results.\n\n3. **Mention the audience**: \"For my CEO\" means executive summary. \"For the data team\" means show the methodology.\n\n4. **Specify what you'll do with it**: \"I need to present this to the board\" vs \"I need clean data for my ML pipeline\" β context shapes the output.\n\n5. **Don't over-specify methods**: Let CellCog choose the right statistical approach. Say what you want to *learn*, not which algorithm to use.\n\n6. **Iterate**: Upload data β get initial analysis β ask follow-up questions β go deeper. CellCog maintains context across messages.\n\n---\n\n## If CellCog is not installed\n\n**Cursor:** Run `/cellcog-setup` to install and authenticate.\n**OpenClaw:** Run `clawhub install cellcog` for SDK setup.\n**Other agents:** `pip install -U cellcog` and set `CELLCOG_API_KEY`. See the **cellcog** skill for SDK reference.\n\nFile v1.0.8:_meta.json\n\n{\n \"ownerId\": \"kn7a96cj9q65e0bhmzahv790en80ffqm\",\n \"slug\": \"data-analysis-cellcog\",\n \"version\": \"1.0.8\",\n \"publishedAt\": 1776036917992\n}\n\nArchive v1.0.7: 2 files, 4721 bytes\n\nFiles: SKILL.md (10175b), _meta.json (140b)\n\nFile v1.0.7:SKILL.md\n\n---\nname: data-cog\ndescription: \"AI data analysis and visualization powered by CellCog. Upload CSVs and get charts, dashboards, statistical reports, and clean data back. Data cleaning, exploratory analysis, hypothesis testing, ML model evaluation, dataset profiling, and data visualization. Full Python access. #1 on DeepResearch Bench (Apr 2026). Analyzes everything, presents it beautifully.\"\nmetadata:\n openclaw:\n emoji: \"π’\"\n os: [darwin, linux, windows]\nauthor: CellCog\nhomepage: https://cellcog.ai\ndependencies: [cellcog]\n---\n\n# Data Cog - Your Data Has Answers, CellCog Finds Them\n\n**Your data has answers. CellCog asks the right questions.** #1 on DeepResearch Bench (Apr 2026) + frontier coding agent.\n\nMost AI tools return code when you ask about data. CellCog returns answers β actual charts, clean datasets, statistical reports, and visual dashboards. Upload messy CSVs with a minimal prompt, and CellCog's coding agent explores your data, finds the patterns, and presents them beautifully. Full Python access for everything from data cleaning to ML model evaluation.\n\n---\n\n## Prerequisites\n\nThis skill requires the `cellcog` skill for SDK setup and API calls.\n\n```bash\nclawhub install cellcog\n```\n\n**Read the cellcog skill first** for SDK setup. This skill shows you what's possible.\n\n**OpenClaw agents (fire-and-forget β recommended for long tasks):**\n```python\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n notify_session_key=\"agent:main:main\", # OpenClaw only\n task_label=\"my-task\",\n chat_mode=\"agent\", # See cellcog skill for all modes\n)\n```\n\n**All other agents (blocks until done):**\n```python\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n)\n```\n\nSee the **cellcog** mothership skill for complete SDK API reference β delivery modes, timeouts, file handling, and more.\n\n---\n\n## What Makes Data-Cog Different\n\n### Code as Tool, Not as Output\n\nOther AI tools give you Python code and say \"run this.\" CellCog **runs the code for you** and delivers the results:\n\n| Other AI Tools | Data-Cog |\n|---------------|----------|\n| \"Here's a pandas script to analyze your data\" | Here are your actual insights with charts |\n| \"Run this matplotlib code to see the chart\" | Here's the chart, annotated with findings |\n| \"This SQL query will find outliers\" | Found 23 outliers, here's what they mean |\n| \"You'll need scikit-learn for this\" | Model trained, here's accuracy and feature importance |\n\nYou upload data. You get answers. The code runs behind the scenes.\n\n---\n\n## What Data Work You Can Do\n\n### Exploratory Data Analysis\n\nUnderstand your data fast:\n\n- **Dataset Profiling**: \"Analyze this CSV β distributions, missing values, outliers, correlations, and data quality summary\"\n- **Pattern Discovery**: \"What patterns and trends exist in this sales data? Surprise me.\"\n- **Anomaly Detection**: \"Find unusual patterns in this server log data β what looks abnormal?\"\n- **Relationship Analysis**: \"What factors most strongly correlate with customer churn in this dataset?\"\n\n**Example prompt:**\n> \"Analyze this dataset:\n> /path/to/customer_data.csv\n> \n> I don't know much about this data yet. Give me:\n> - Overview: rows, columns, data types, missing values\n> - Key distributions and summary statistics\n> - Most interesting correlations\n> - Any outliers or data quality issues\n> - 3-5 insights that jump out\n> \n> Present findings as an interactive HTML report with charts.\"\n\n### Data Cleaning & Transformation\n\nWrangle messy data into shape:\n\n- **Clean Messy Data**: \"Clean this CSV β fix inconsistent date formats, handle missing values, remove duplicates, standardize column names\"\n- **Data Transformation**: \"Pivot this transaction data into a monthly summary by product category\"\n- **Data Merging**: \"Join these three CSV files on customer_id and create a unified dataset\"\n- **Feature Engineering**: \"Create useful features from this raw data for predicting house prices\"\n\n**Example prompt:**\n> \"Clean and transform this dataset:\n> /path/to/messy_data.csv\n> \n> Issues I know about:\n> - Dates are in mixed formats (MM/DD/YYYY and YYYY-MM-DD)\n> - 'Revenue' column has some values with $ signs and commas\n> - Duplicate rows exist\n> - Missing values in 'Region' column\n> \n> Clean it up and give me back a clean CSV plus a summary of what you changed.\"\n\n### Statistical Analysis\n\nRigorous analysis with real numbers:\n\n- **Hypothesis Testing**: \"Is there a statistically significant difference in conversion rates between our A and B variants?\"\n- **Regression Analysis**: \"What factors predict employee salary in this HR dataset? Build a regression model.\"\n- **Time Series Analysis**: \"Analyze this monthly revenue data β trend, seasonality, and forecast next 6 months\"\n- **Cohort Analysis**: \"Create a cohort analysis showing user retention by signup month\"\n\n**Example prompt:**\n> \"I ran an A/B test on our checkout page:\n> /path/to/ab_test_results.csv\n> \n> Columns: user_id, variant (A or B), converted (0/1), revenue, timestamp\n> \n> Tell me:\n> - Is variant B statistically better? (p-value, confidence interval)\n> - Conversion rate difference\n> - Revenue per user difference\n> - Sample size adequacy check\n> - My recommendation: ship B or keep testing?\n> \n> Present with clear charts and a plain-English conclusion.\"\n\n### Visualization & Reporting\n\nTurn data into visual stories:\n\n- **Chart Generation**: \"Create a set of charts showing our quarterly performance from this data\"\n- **Dashboard Reports**: \"Build an interactive dashboard from this sales dataset with filters by region and product\"\n- **Presentation-Ready Visuals**: \"Create publication-quality charts from this research data\"\n- **Comparison Visuals**: \"Visualize how our metrics compare to industry benchmarks\"\n\n### Machine Learning\n\nApplied ML without the setup:\n\n- **Classification**: \"Predict which customers will churn based on this dataset β train a model, show feature importance\"\n- **Clustering**: \"Segment these customers into groups based on behavior β how many natural clusters exist?\"\n- **Forecasting**: \"Forecast next quarter's sales using this historical data\"\n- **Model Evaluation**: \"I trained a model β here are the predictions. Evaluate: accuracy, precision, recall, confusion matrix, ROC curve\"\n\n**Example prompt:**\n> \"Predict customer churn from this dataset:\n> /path/to/customer_features.csv\n> \n> Target column: 'churned'\n> \n> - Train a model, try at least 2 algorithms\n> - Show feature importance β what drives churn?\n> - Confusion matrix and ROC curve\n> - Plain-English summary: 'The top 3 reasons customers churn are...'\n> - Actionable recommendations based on findings\n> \n> I want insights, not just metrics.\"\n\n---\n\n## Supported Data Formats\n\n| Format | How to Send |\n|--------|-------------|\n| **CSV** | Upload via SHOW_FILE |\n| **Excel (XLSX)** | Upload via SHOW_FILE |\n| **JSON** | Upload via SHOW_FILE |\n| **Parquet** | Upload via SHOW_FILE |\n| **SQL exports** | Upload the dump via SHOW_FILE |\n| **Inline data** | Describe small datasets directly in prompt |\n\n---\n\n## Output Formats\n\n| Format | Best For |\n|--------|----------|\n| **Interactive HTML Dashboard** | Explorable charts, filters, drill-downs |\n| **PDF Report** | Shareable analysis reports with charts and findings |\n| **Clean CSV/XLSX** | Cleaned or transformed data files for downstream use |\n| **Markdown** | Quick insights for integration into docs |\n\n---\n\n## Chat Mode for Data\n\n| Scenario | Recommended Mode |\n|----------|------------------|\n| Quick data cleaning, simple charts, basic statistics | `\"agent\"` |\n| Deep analysis with multiple techniques, ML modeling, comprehensive reports | `\"agent team\"` |\n\n**Use `\"agent\"` for most data work.** Data cleaning, EDA, chart generation, and standard statistical analysis execute well in agent mode.\n\n**Use `\"agent team\"` for complex analytical projects** β multi-technique analysis, ML model comparisons, or when you need deep domain reasoning about what the data means.\n\n---\n\n## Example Prompts\n\n**Minimal prompt, maximum insight:**\n> \"Analyze this:\n> /path/to/data.csv\n> \n> Tell me everything interesting.\"\n\nThat's it. CellCog's coding agent will profile the data, run exploratory analysis, find patterns, and present findings with charts. You don't need to know what to ask β the agent figures it out.\n\n**Business analysis:**\n> \"Analyze our e-commerce data:\n> /path/to/orders.csv\n> \n> I need:\n> - Revenue trends (daily, weekly, monthly)\n> - Best and worst performing products\n> - Customer purchase frequency distribution\n> - Average order value trends\n> - Seasonal patterns\n> - Top 5 actionable insights for growing revenue\n> \n> Interactive HTML dashboard with all charts.\"\n\n**Research data analysis:**\n> \"Analyze this survey data from 500 respondents:\n> /path/to/survey.csv\n> \n> Research questions:\n> 1. Is there a significant relationship between age group and product preference?\n> 2. Do satisfaction scores differ by region? (ANOVA)\n> 3. What factors best predict likelihood to recommend? (regression)\n> \n> Include: statistical tests, p-values, effect sizes, and publication-ready charts.\n> PDF report format.\"\n\n---\n\n## Tips for Better Data Analysis\n\n1. **Just upload and ask**: You don't need to describe every column. CellCog reads the data and figures out what's there.\n\n2. **State your question**: \"What drives churn?\" is more focused than \"Analyze this data.\" Both work, but the first gets faster results.\n\n3. **Mention the audience**: \"For my CEO\" means executive summary. \"For the data team\" means show the methodology.\n\n4. **Specify what you'll do with it**: \"I need to present this to the board\" vs \"I need clean data for my ML pipeline\" β context shapes the output.\n\n5. **Don't over-specify methods**: Let CellCog choose the right statistical approach. Say what you want to *learn*, not which algorithm to use.\n\n6. **Iterate**: Upload data β get initial analysis β ask follow-up questions β go deeper. CellCog maintains context across messages.\n\nFile v1.0.7:_meta.json\n\n{\n \"ownerId\": \"kn7a96cj9q65e0bhmzahv790en80ffqm\",\n \"slug\": \"data-analysis-cellcog\",\n \"version\": \"1.0.7\",\n \"publishedAt\": 1775627871920\n}","readmeExcerpt":"Skill: Data Analysis Owner: cellcog Summary: AI data analysis and visualization powered by CellCog. Data cleaning, exploratory analysis, hypothesis testing, statistical reports, ML model evaluation, dataset profiling, charts, dashboards. Upload CSVs and get analysis back with full Python access. Tags: latest:1.0.16 Version history: v1.0.16 | 2026-08-24T02:01:10.439Z | user Content updated. v1.0.15 | 2026-08-24T01:45:","codeSnippets":[],"executableExamples":[{"language":"python","snippet":"result = client.create_chat(\n prompt=\"[your task prompt]\",\n notify_session_key=\"agent:main:main\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n chat_tier=\"max\",\n)"},{"language":"python","snippet":"from cellcog import CellCogClient\nclient = CellCogClient(agent_provider=\"openclaw|cursor|claude-code|codex|...\")\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n chat_tier=\"max\",\n)\nprint(result[\"message\"])"},{"language":"python","snippet":"result = client.create_chat(\n prompt=\"[your task prompt]\",\n notify_session_key=\"agent:main:main\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n chat_tier=\"max\",\n)"},{"language":"python","snippet":"from cellcog import CellCogClient\nclient = CellCogClient(agent_provider=\"openclaw|cursor|claude-code|codex|...\")\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n chat_tier=\"max\",\n)\nprint(result[\"message\"])"},{"language":"python","snippet":"result = client.create_chat(\n prompt=\"[your task prompt]\",\n notify_session_key=\"agent:main:main\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n)"},{"language":"python","snippet":"from cellcog import CellCogClient\nclient = CellCogClient(agent_provider=\"openclaw|cursor|claude-code|codex|...\")\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n)\nprint(result[\"message\"])"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: data-analysis-cellcog\ndescription: \"AI data analysis and visualization powered by CellCog. Data cleaning, exploratory analysis, hypothesis testing, statistical reports, ML model evaluation, dataset profiling, charts, dashboards. Upload CSVs and get analysis back with full Python access.\"\nmetadata:\n openclaw:\n emoji: \"π’\"\n os: [darwin, linux, windows]\n requires:\n bins: [python3]\n env: [CELLCOG_API_KEY]\nauthor: CellCog\nhomepage: https://cellcog.ai\ndependencies: [cellcog]\n---\n# Data Analysis - Your Data Has Answers, CellCog Finds Them\n\nData analysis and visualization from uploaded files.\n\nMost AI tools return code when you ask about data. CellCog returns answers β actual charts, clean datasets, statistical reports, and visual dashboards. Upload messy CSVs with a minimal prompt, and CellCog's coding agent explores your data, finds the patterns, and presents them beautifully. Full Python access for everything from data cleaning to ML model evaluation.\n\n## How to Use\n\nFor your first CellCog task in a session, read the **cellcog** skill for the full SDK reference β file handling, chat modes, timeouts, and more.\n\n**OpenClaw (fire-and-forget):**\n```python\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n notify_session_key=\"agent:main:main\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n chat_tier=\"max\",\n)\n```\n\n**All agents except OpenClaw (blocks until done):**\n```python\nfrom cellcog import CellCogClient\nclient = CellCogClient(agent_provider=\"openclaw|cursor|claude-code|codex|...\")\nresult = client.create_chat(\n prompt=\"[your task prompt]\",\n task_label=\"my-task\",\n chat_mode=\"agent\",\n chat_tier=\"max\",\n)\nprint(result[\"message\"])\n```\n\n\n---\n\n## What Makes Data-Cog Different\n\n### Code as Tool, Not as Output\n\nOther AI tools give you Python code and say \"run this.\" CellCog **runs the code for you** and delivers the results:\n\n| Other AI Tools | Data-Cog |\n|---------------|----------|\n| \"Here's a pandas script to analyze your data\" | Here are your actual insights with charts |\n| \"Run this matplotlib code to see the chart\" | Here's the chart, annotated with findings |\n| \"This SQL query will find outliers\" | Found 23 outliers, here's what they mean |\n| \"You'll need scikit-learn for this\" | Model trained, here's accuracy and feature importance |\n\nYou upload data. You get answers. The code runs behind the scenes.\n\n---\n\n## What Data Work You Can Do\n\n### Exploratory Data Analysis\n\nUnderstand your data fast:\n\n- **Dataset Profiling**: \"Analyze this CSV β distributions, missing values, outliers, correlations, and data quality summary\"\n- **Pattern Discovery**: \"What patterns and trends exist in this sales data? Surprise me.\"\n- **Anomaly Detection**: \"Find unusual patterns in this server log data β what looks abnormal?\"\n- **Relationship Analysis**: \"What factors most strongly correlate with customer churn in this dataset?\"\n\n**Example prompt:**\n> \"Analyze this dataset:\n> /path/to/customer_data.csv