CLIO Geo

Check the status of a Slurm job by its ID.

Remove duplicates, detect outliers via IQR/Z-score, and optimize data types in a single pass.

Close the current HDF5 file. Returns: Status message

Compare two Darshan log files to identify performance differences between runs.

Compress a file using gzip. Returns original/compressed sizes and compression ratio.

Compute the surface area of the active dataset (must be a surface mesh). Returns: Status message with area value.

Compute per-trace amplitude statistics (min, max, mean, std, peak_abs) plus header metadata (npts, delta_s, begin_s, end_s) for SAC traces in a file or archive. Read-only; bounded by max_traces.

Configure a package in a Jarvis-CD pipeline.

Toggle volume rendering visibility for the active source. Args: enable: Whether to show (True) or hide (False) volume rendering. Returns: Status message with source name.

Compute correlation matrices (Pearson, Spearman, or Kendall) with significance testing and strong-correlation detection.

Create a slice plane through the loaded volume data. Args: origin_x, origin_y, origin_z: Slice origin coordinates (defaults to data center). normal_x, normal_y, normal_z: Normal vector for the slice plane (default [0, 0, 1]). Returns: Status message with pipeline name.

Create a geometric source (Sphere, Cone, Cylinder, Plane, or Box). Args: source_type: Type of source to create. Returns: Status message with source name.

Create a new Jarvis-CD pipeline environment.

Get schema, column types, and summary statistics for a CSV or Excel file.

Release a Slurm node allocation by canceling it.

Decompress a gzip-compressed (.gz) file back to its original form.

Analyze a Digital Elevation Model grid for elevation, slope, aspect, and site suitability. Accepts CSV numeric grids, NPY, and NPZ (with a 'dem' array); GeoTIFF requires the optional rasterio extra. Returns grid shape, summary statistics, suitability counts, and representative suitable cells.

Destroy a pipeline environment and clean up files.

Detect patterns in log files including anomalies and error clusters.

Download multiple PDFs concurrently with rate limiting.

Download the PDF of a paper from ArXiv.

Edit the opacity transfer function for a scalar field. Args: field_name: The scalar field to modify. opacity_points: List of dicts like [{"value": 0.0, "alpha": 0.0}, ...]. Returns: Status message.

Export dataset to various formats with user format selection. Args: path: Path to dataset within file output_path: Optional output file path ctx: Context for elicitation Returns: Export summary

Export a structured snapshot of a Jarvis-CD pipeline.

Export search results to BibTeX format for citation management.

Export log entries to CSV format with structured columns.

Export log processing results to JSON format.

Export log entries to plain text format.

Compute the overall bounding box [min_lon, min_lat, max_lon, max_lat] of all features in a GeoJSON document. Accepts a file path or inline GeoJSON. Returns the bbox (or null when there are no coordinates) and the feature count.

Fetch an HTTP(S) URL with a streamed size cap and timeout, convert HTML to Markdown, and return the content inline or (to_file=True) write it to a local file and return its path.

Filter log entries by keywords with support for multiple keywords and logical operations.

Filter log entries by log level (ERROR, WARN, INFO, DEBUG, etc.).

Filter log entries by time range using start and end timestamps.

Filter rows using comparison, membership, pattern-matching, and null-check operators across multiple columns.

Filter log entries based on multiple conditions with logical operations.

Filter/rank any table of points by great-circle distance to a center. Reads a CSV (or GeoJSON points) of locations, computes the haversine distance from (center_lat, center_lon) to each row, and returns the rows within radius_km sorted ascending by distance, each annotated with distance_km. Latitude/longitude columns are auto-detected from common names (latitude/lat, longitude/lon/long) when not given. Domain-neutral: works for sensors, sites, cities, samples, or any lon/lat table.

Find datasets with similar characteristics to a reference dataset with AI analysis. Args: reference_path: Path to reference dataset similarity_threshold: Similarity threshold (0.0 to 1.0) ctx: Context for LLM sampling Returns: List of similar datasets with similarity scores and AI insights

Find papers similar to a reference paper based on categories and keywords.

Create streamlines from a vector volume using the StreamTracer filter. Args: seed_point_number: Number of seed points to generate. vector_field: Vector field name (auto-detected if None). integration_direction: 'FORWARD', 'BACKWARD', or 'BOTH'. max_steps: Maximum integration steps. initial_step: Initial step length. maximum_step: Maximum streamline length. Returns: Status message with tube name.

Generate a comprehensive I/O summary report with findings and recommendations.

Create an isosurface visualization of the active source at the given isovalue. Args: value: Isovalue. field: Optional field name to contour by. Returns: Status message with filter name.

Generate a summary report of log processing results with statistics.

Look up a free-text place name or location and return real coordinates from OpenStreetMap Nominatim (a lookup, not a model guess). Each match carries lat/lon, a [min_lon, min_lat, max_lon, max_lat] bbox, type, importance, and a provenance source so the region can be grounded and cited.

List pipeline source names, optionally filtered by type. Args: source_type: Filter by type (e.g., 'Sphere', 'Contour'). None returns all. Returns: Formatted list of source names.

Check the status of a Slurm node allocation.

List available data arrays in the active source. Returns: List of available arrays.

Get a dataset or group by path. Args: path: Path to object within file Returns: Object information

Get chunk information for a dataset. Args: path: Path to dataset Returns: Chunk configuration

Preview values from a specific column with pagination.

Get CPU specifications, core counts, frequencies, and per-core usage.

Retrieve detailed metadata for a specific dataset by ID or name.

Get disk partitions, usage statistics, and I/O counters.

Get the data type of a dataset. Args: path: Path to dataset Returns: Dataset dtype

Get the current file's path. Returns: File path

Get GPU model, memory, temperature, and utilization via nvidia-smi/rocm-smi.

Compute histogram data for a field in the active source. Args: field: Field name (auto-selected if only one exists). num_bins: Number of bins (default: 256). data_location: 'POINTS' or 'CELLS'. Returns: Formatted histogram data.

Extract I/O performance metrics including bandwidth, IOPS, and request sizes.

Get detailed information about a specific Slurm job.

Retrieve stdout or stderr output from a Slurm job.

Get job-level summary from a Darshan log including runtime, process count, and I/O volume.

Get RAM and swap capacity, usage percentages, and availability.

Get the current file's access mode. Returns: File mode

Get network interfaces, IP addresses, and I/O statistics.

Get information about Slurm cluster nodes and their resources.

Get detailed information about a specific ArXiv paper by ID.

Get the direct PDF URL for a paper without downloading.

Get real-time CPU, memory, disk, and network performance metrics.

Get the current visualization pipeline structure. Returns: Description of the current pipeline.

Retrieve the configuration of a specific package in a pipeline.

Get running processes with CPU, memory, and status details.

Get Slurm queue status and partition information.

Get recent papers from a specific ArXiv category.

Collect hardware info from a remote node via SSH. Supports component filtering.

Get temperature, fan speed, and battery sensor readings.

Get the shape of a dataset. Args: path: Path to dataset Returns: Dataset shape

Get the size of a dataset. Args: path: Path to dataset Returns: Dataset size

Get Slurm cluster configuration, partitions, and resource availability.

Get OS details, hostname, uptime, and active users.

Generate timeline analysis showing I/O activity over time and temporal patterns.

Group data by columns and apply aggregations (sum, mean, count, min, max, std, median) with optional pre-filter.

Detect, impute, or remove missing values using strategies like mean/median/mode fill, forward/backward fill, or interpolation.

Parallel statistics computation across multiple datasets. Args: paths: Comma-separated dataset paths or JSON array stats: Comma-separated stats to compute (default: mean,std,min,max,sum,count) ctx: Context for progress reporting Returns: Aggregate statistics summary

Read multiple datasets in parallel. Args: paths: Comma-separated dataset paths or JSON array slice_spec: Optional slice specification ctx: Context for progress reporting Returns: Batch read summary

Fast multi-file scanning with parallel processing. Args: directory: Directory to scan pattern: File pattern (default: *.h5) ctx: Context for progress reporting Returns: Scan summary with file metadata

Stream large datasets efficiently with memory management. Args: path: Path to dataset chunk_size: Number of elements per chunk max_chunks: Maximum number of chunks to process ctx: Context for progress reporting Returns: Stream processing summary with statistics

Verify server health and hardware monitoring capability status.

Create a correlation heatmap from numeric columns in CSV or Excel data.

Create a histogram from CSV or Excel data showing value distribution.

Run statistical hypothesis tests (t-test, chi-square, ANOVA, normality, Mann-Whitney) with p-values and effect sizes.

Identify I/O performance bottlenecks by analyzing access patterns and operations.

Inspect a staged SAC file or TAR archive and summarize its SAC waveform members: count, a sample of member names and sizes, and the inferred stations and phases. Read-only; a good first step before computing statistics or plotting.

Reads global or variable-specific attributes from a BP5 file. The 'variable_name' is optional.

Inspect a GeoJSON document and report its geometry types and counts, feature count, property keys (schema), bounding box [min_lon, min_lat, max_lon, max_lat], CRS if present, and total vertex count. Accepts a file path or inline GeoJSON.

Inspects variables in a BP5 file, returning type, shape, and steps. Optionally filters by variable name.

Inspects a specific variable at a given step in a BP5 file. All parameters are required.

Add and configure a package-backed step in a JARVIS pipeline. First use jarvis_describe(target='package') for the selected package; config keys must use its canonical setting names exactly, except for aliases explicitly listed there. Explicit null is accepted only when that setting reports nullable=true; omit a setting to use its declared default. Only settings marked agent-visible by the package are accepted here. User-level step configuration is always validated and cannot be bypassed.

Create a JARVIS pipeline. Optionally pass execution intent such as local, cluster, or hostfile mode; backend details are resolved where the MCP server runs.

Describe JARVIS packages, one package, a pipeline, or one pipeline step. For a named application, first use target='package' with its unique short name or fully qualified package name. Use target='package_search' for bounded discovery, then describe the selected canonical name. A package description returns package-owned configuration metadata and, when supported, a versioned deployment contract covering execution profiles, runtime requirements, readiness, and configuration rules. Package settings include only semantic parameters explicitly marked agent-visible. A setting may include a version…

Edit or remove a step in a JARVIS pipeline. Use operation='edit' with config, or operation='remove' without config.

Query one JARVIS execution handle, durable lifecycle record, and runtime metadata. Progress is included by default and can be omitted. Set include_service_runtimes=true to include execution-owned network services; authenticated services expose only a non-secret bearer token SHA-256 fingerprint. Set artifacts to {} or filters to include one bounded artifact page; omit artifacts to avoid querying the artifact manifest.

Start a configured JARVIS pipeline and return its durable execution handle without waiting for workload completion. Optional execution intent selects local, cluster, or hostfile mode without exposing scheduler internals. For each runtime resolved with Spack, copy spack_locate.output.load_spec unchanged into one element of jarvis_run.input.spack_specs; do not derive an executable path from the Spack prefix. JARVIS resolves those specs into a filtered environment and persists it before direct or scheduler execution. Use jarvis_get_execution with the returned pipeline_id and execution_id to query…

Add a repository to JarvisManager.

Bootstrap Jarvis config from a machine template.

List available bootstrap machine templates.

Change current Jarvis pipeline context.

Construct a package skeleton in JarvisManager.

Initialize JarvisManager config directories.

Get repository info from JarvisManager.

Build or rebuild the resource graph with a net sleep interval.

Modify the resource graph using a net sleep interval.

Print the current resource graph frames.

List all existing Jarvis pipelines.

List all Jarvis repositories.

Load existing JarvisManager configuration.

Promote a repository in JarvisManager.

Remove a repository from JarvisManager.

Reset JarvisManager (destroy all pipelines and data).

Save current JarvisManager configuration.

Set hostfile path for JarvisManager.

Create a line plot from CSV or Excel data with customizable styling.

List all attributes of an object. Args: path: Path to object Returns: JSON dict of attributes

List all registered HDF5 files with resource URIs for Claude Code @ mentions. Returns: List of available files with resource URIs

Lists all BP5 files in a given directory. The 'directory' parameter must be an absolute path.

List all available commands in this ParaView MCP server. Returns: List of available commands.

List keys in a group. Args: path: Path to group (default: root) Returns: JSON array of keys

List organizations available in the National Data Platform.

List Slurm jobs with optional filtering by user and state.

Load and parse a Darshan log file to extract I/O performance metrics and metadata.

Load and parse data from CSV, Excel, JSON, Parquet, or HDF5 files with optional column selection and row limiting.

Load an existing Jarvis-CD pipeline environment.

Load scientific datasets (VTK, EXODUS, CSV, RAW, BP5) into ParaView with automatic format detection.

Join two datasets using inner, outer, left, or right joins on specified key columns.

Search for available modules, optionally filtered by name pattern.

List all currently loaded environment modules.

Load one or more environment modules into the current session.

Restore a previously saved module collection.

Save currently loaded modules as a named collection.

List all saved module collections.

Display detailed information about a specific module.

Search the entire module tree comprehensively for matching modules.

Swap one module for another atomically.

Unload one or more currently loaded modules from the environment.

Open an HDF5 file for operations. Args: path: Path to HDF5 file mode: File access mode ('r', 'r+', 'w', 'a') Returns: Success message with file info

Suggest better approaches for data access based on usage patterns. Args: dataset_path: Path to dataset access_pattern: Access pattern (sequential, random, batch) Returns: Optimization recommendations

Analyze and reduce DataFrame memory usage through automatic dtype optimization and chunked-processing recommendations.

Sort large log files using parallel processing with chunked approach.

Create pivot tables with configurable row index, column headers, value columns, and aggregation function.

Create a 'Plot Over Line' filter to sample data between two points. Args: point1: Start [x, y, z] coordinates (defaults to data bounds). point2: End [x, y, z] coordinates (defaults to data bounds). resolution: Number of sample points (default: 100). Returns: Status message.

Render the three-panel earthquake-sequence figure from a saved catalog: (1) an epicenter map sized by magnitude and coloured by time, (2) the Gutenberg-Richter magnitude-frequency distribution with an optional b-value fit line, and (3) the cumulative count over time. Writes a PNG and returns its path; pass mc/b_value to draw the G-R fit line.

Create a multi-series line chart PNG from one or more y columns of a CSV or Excel file, auto-detecting a time, numeric, or categorical x axis.

Plot selected SAC traces from a file or archive to a PNG artifact. Traces are amplitude-normalized and vertically offset. Writes a file; returns the output path, plotted member names, and render duration.

Read an x/y/z point cloud and grid it into a DEM-like surface by averaging z per cell. Accepts CSV with x,y,z columns, NPY, and NPZ; LAS/LAZ requires the optional laspy extra. Optionally writes the gridded surface to a CSV DEM for downstream dem_terrain analysis. Returns point/grid stats and bounds.

Spatial overlap: return which GeoJSON points fall within (optionally buffered) GeoJSON polygons — e.g. which AirNow monitors lie inside the smoke footprint. Accepts inline GeoJSON or file paths. Returns the matched points with their properties and a matched_count.

Quickly profile a CSV file: row/column counts, per-column dtype, null counts, and min/max/mean for numeric columns.

Generate a full dataset profile: shape, types, missing values, distributions, quality checks, and optional correlations.

Query an ArcGIS FeatureServer layer (with optional lon/lat bbox and where clause) and write the returned features to a local GeoJSON file. The saved FeatureCollection can be fed directly into render_feature_map, points_in_polygons, or bounding_box.

Read an attribute from an object. Args: path: Path to object name: Attribute name Returns: Attribute value

Read an entire dataset with efficient chunked reading for large datasets. Args: path: Path to dataset within file Returns: Dataset description

Read a portion of a dataset with slicing. Args: path: Path to dataset within file start: Starting indices as comma-separated string (e.g., "0,0,0") count: Number of elements as comma-separated string (e.g., "10,10,10") Returns: Partial dataset description

Read a row slice from a Parquet file with optional column projection and filtering.

Reads a named variable at a specific step from a BP5 file. All parameters are required.

Log a user message and LLM response to the active ChronoLog story.

Re-scan client roots and update available HDF5 resources. FastMCP automatically sends notifications/resources/list_changed to clients. Returns: Summary of refreshed resources

Delete a package through JARVIS-CD's destructive removal API. Fails explicitly when the installed JARVIS-CD only supports non-destructive unlinking.

Render one or more GeoJSON layers (polygons/lines/points) onto a single map PNG with an optional basemap. Each layer accepts a style with fixed colors or data-driven coloring (color_by + category_colors, an 'epa_aqi' AQI scale, or a matplotlib colormap). Returns the output path and bounds.

Reset the camera to show all data in the viewport. Returns: Status message.

Retrieve logged records from a chronicle and story, with optional time filtering.

Rotate the camera by azimuth and elevation angles in degrees. Args: azimuth: Rotation around vertical axis. elevation: Rotation around horizontal axis. Returns: Status message.

Execute a Jarvis-CD pipeline end-to-end.

Save the active contour or surface as an STL file in the data directory. Args: stl_filename: The STL file name to use, defaults to 'contour.stl'. Returns: Status message.

Save data to CSV, Excel, JSON, Parquet, or HDF5 with auto-detected format and optional index inclusion.

Create a scatter plot from CSV or Excel data for correlation analysis.

Return one exact operator catalog record plus a top-level dataset_descriptor. Pass dataset_descriptor unchanged as jarvis_add_step config.dataset_descriptor; do not pass the surrounding dataset record.

Search operator-registered scientific datasets and return bounded intrinsic summaries.

Search the web via a configurable provider (keyless DuckDuckGo by default; optional BYO-key Brave or Tavily) and return ranked results.

Search ArXiv for papers by category or topic.

Search ArXiv papers by abstract keywords.

Search ArXiv papers by subject classification.

Search ArXiv papers by title keywords.

Search for datasets in the NDP using term-based or field-specific criteria.

Search ArXiv papers within a specific date range.

Search ArXiv papers by author name.

Set the active pipeline object by its registered name. Args: name: The pipeline source name (e.g., 'Contour1'). Returns: Status message.

Set a custom color transfer function for volume rendering. Args: field_name: The field/array name in ParaView. color_points: List of dicts: {"value": float, "rgb": [r, g, b]}. Returns: Status message.

Apply a predefined color map preset (e.g., Viridis, Plasma, Cool to Warm). Args: preset_name: Name of the color map preset. Returns: Status message.

Set the representation type for the active source (Surface, Wireframe, Points, etc.). Args: rep_type: Representation type. Returns: Status message.

Capture a screenshot with inline preview using temporary files.

Request destructive cancellation of one Slurm job. confirm_job_id must exactly repeat job_id; omission or mismatch is rejected without calling scancel.

Inspect bounded Slurm partition and queue records in one snapshot. Node details are excluded by default and bounded when requested.

Describe one scheduler-native Slurm job: lifecycle state, terminality, scheduler properties, and optional bounded stdout/stderr tails.

List a bounded number of Slurm jobs with optional user, state, and partition filters. Returns native IDs and explicit truncation state.

Submit one Slurm job or array and return its scheduler-native job ID. Set array only for an array submission.

Sort log file lines by timestamps in YYYY-MM-DD HH:MM:SS format.

Admin diagnostic: return a filtered structured environment for installed specs. This does not mutate the MCP server or any later process.

List installed Spack packages matching an optional constraint. No matches is a successful result with count=0 and packages=[].

Install one Spack spec with explicit reusable or fresh concretization and verify that a matching install is observable.

Resolve one unique installed Spack spec. Copy spack_locate.output.load_spec unchanged into one element of jarvis_run.input.spack_specs; do not derive or pass an executable path from the returned prefix. An absent package returns the structured not_installed error.

Download/stage an HTTP(S) or OSDF/Pelican dataset resource to a local file and return its local_path, size, and content-type.

Connect to ChronoLog, create a chronicle, and acquire a story handle.

Compute descriptive statistics, distribution analysis, and outlier detection for numerical and categorical columns.

Release the story handle and disconnect from ChronoLog.

Submit a Slurm array job for parallel task execution.

Submit a job script to the Slurm scheduler with resource requirements.

Suggest interesting data to explore next based on current location with AI recommendations. Args: current_path: Current path (default: root) ctx: Context for LLM sampling Returns: Exploration suggestions with interest scores and AI recommendations

Produce a compact human-readable summary of a GeoJSON document: counts per geometry type, bounding box, property keys, and a few sample feature property sets. Accepts a file path or inline GeoJSON.

Return Parquet schema, row count, and file size.

Capture a screenshot of the current ParaView viewport and save it as a timestamped PNG.

Resample, compute rolling statistics, create lag features, or difference a time series.

Toggle visibility for the active source. Args: enable: Whether to show (True) or hide (False) the active source. Returns: Status message with source name.

Unlink a package from a pipeline (preserve files).

Re-apply environment and configuration to every package in a pipeline.

Validate columns against rules for min/max range, data type, nullability, uniqueness, and regex patterns.

Validate the structural well-formedness of a GeoJSON document: that the top-level type is recognized and every geometry's type and coordinates are well-formed (correct nesting depth, finite numeric positions). Returns {valid, errors}.

Visit all nodes recursively. Args: callback_fn: Callback function name (currently collects all paths) Returns: JSON array of all paths and types

Apply a 'Warp By Vector' filter to the active source. Args: vector_field: Vector field name (auto-detected if None). scale_factor: Scale factor for the warp. Returns: Status message.