CatchAll MCP Server

Add existing entities to a dataset.

Create a new dataset. Datasets are collections of entities (companies/people). Connect a dataset to a job via `submit_query(connected_dataset_ids=[...])` to narrow retrieval scope.

Create a new dataset by uploading a CSV file. The CSV must have at least a `name` column. For meaningful entity enrichment each row should also include a `domain` column or a `description` column (or both) — a row with only a name is accepted but produces lower-quality enrichment. Additional columns are mapped to entity attributes. Max file size is plan-dependent. To add CSV rows to an existing dataset, use `append_csv_to_dataset` instead.

Permanently delete a dataset. The entities the dataset referenced are not deleted; only the dataset and its entity associations are removed.

Get a single dataset's details.

Get the status history of a dataset (e.g. its enrichment progress over time).

List the entities contained in a dataset.

Remove entities from a dataset (the entities themselves are not deleted).

Update a dataset's name and/or description.

Add one or more resources to a project. Webhooks are first-class project resources: a webhook can belong to several projects at the same time, and deleting a project only detaches its webhooks — it never deletes them.

Create a new project. Projects group related resources (jobs, monitors, datasets, monitor_groups) so you can organize work and filter listings by `project_id`.

Delete a project. By default the project's resources (jobs, monitors, etc.) are detached but kept. Set `delete_resources=true` to also delete the contained jobs, monitors, datasets, and monitor groups. Webhooks are the exception: they are never deleted by this operation — an attached webhook is only detached from the project and keeps working (it may belong to other projects or resources independently of this one).

Get a single project's details.

Get a project's resource overview (counts grouped by resource type and status).

List the resources contained in a project.

Remove a single resource from a project. This detaches the resource from the project without deleting the resource itself (e.g. removing a webhook only ends its membership in this project; the webhook keeps existing and stays attached to any other projects).

Update a project's name and/or description. Only the fields you provide are changed.

Create a new webhook endpoint. Use when: - You want to register a URL to receive job or monitor result deliveries. - You need a webhook_id to attach to a monitor (via webhook_ids) or a job submission. - You want the webhook associated with a project from the start (pass `project_id`).

Permanently delete a webhook endpoint. Use when: - You want to remove a webhook from your account.

Retrieve the full configuration of a specific webhook. Use when: - You want to inspect a webhook's URL, method, headers, or status by its ID.

Get webhook delivery history, either for a resource or for a webhook. Query in exactly one of two modes: - By resource: pass `resource_type` + `resource_id` to see deliveries made for a specific job/monitor/monitor_group. - By webhook: pass `webhook_id` to see every delivery made through one webhook — including manual test deliveries (from `test_webhook`), which are not tied to a job or monitor and only appear in this mode.

List the resources mapped to a webhook. Use when: - You want to see which jobs/monitors a webhook is attached to.

Unmap a resource from a webhook. Use when: - You want to stop a webhook from firing for a specific job or monitor.

Update an existing webhook's configuration. Use when: - You want to change a webhook's URL, method, headers, or other settings. - You want to enable or disable a webhook (set `is_active`). - Only the fields you provide are updated; omitted fields remain unchanged.

Create a recurring monitor from a completed job. Monitors re-run a job's query on a schedule. Use the explore -> refine -> automate pattern: submit a job, refine until results match, then create a monitor. The schedule is defined in natural language (e.g., 'every day at 9 AM EST'). Always include a timezone (in the schedule text or via the `timezone` arg). API-enforced constraints apply: - If `backfill=true`, reference job end_date must be within the last 7 days - If `backfill=false`, reference job age does not matter - Minimum schedule frequency depends on your plan Webhooks are now centralized: register them with `create_webhook`, then pass their IDs here via `webhook_ids` (there is no inline webhook config anymore).

Permanently delete a monitor and stop its scheduled runs. Use when: - You want to remove a monitor entirely (use `disable_monitor` to only pause it).

Get the status history of a monitor. Use when: - You want to see the timeline of a monitor's state changes (e.g. active, disabled, errored) and any related details.

List all jobs spawned by a monitor. Returns the history of scheduled runs for a monitor.

Update a monitor's schedule, timezone, webhook assignments, and per-run limit. Note: reference_job_id cannot be modified through this endpoint. Webhooks are centralized — pass webhook IDs (from `create_webhook`/`list_webhooks`).

Create a single entity (a company or person). ``name`` is required plus at least one identifying field: either ``description`` or ``additional_attributes.company_attributes.domain``.

Permanently delete an entity.

Get a single entity's details.

Update an entity's name, description, external_entity_id, and/or attributes.

Download a job's results as a CSV file. Use when: - You want the full job output as a CSV for offline analysis or export. - Prefer this over `pull_results` when the consumer needs spreadsheet/CSV format.

Download the latest monitor run's results as a CSV file. Use when: - You want the most recent monitor run output as a CSV for offline analysis or export. - Prefer this over `pull_monitor_results` when the consumer needs spreadsheet/CSV format.

Retrieve the latest results from a monitor. Returns the most recent run's results including run_info, records, and all_records.

Retrieve the results of a job. Can be called before completion for partial results, or after completion for the full set. Returns clustered, validated, and enriched web results. While job status is active, call this repeatedly (typically page=1) to refresh partial output. When job reaches completed, iterate all pages. If job fails, call once more to capture any partial output.

Create multiple entities in one call.

List your entities.

Permanently delete a job and its results. Use when: - You want to remove a job you no longer need from your account.

Check the status of a submitted job. Call this after submit_query to see if your job is ready. Status progression: submitted -> analyzing -> fetching -> clustering -> enriching -> completed/failed IMPORTANT: Jobs take several minutes to process. First check after ~1-2 minutes, then poll every 30-60 seconds. Broad searches can take 10-30+ minutes; for long jobs, poll every 60-120 seconds. Do NOT call this tool in a tight loop. Stop polling when status is `completed` or `failed`. Treat `submitted`, `analyzing`, `fetching`, `clustering`, and `enriching` as active states and continue polling. You don't need to wait for completion to pull results. Partial results are available during `enriching` — call pull_results after ~2 minutes, then poll status every 30-60 seconds and pull again for fresher results. Do not stop pulling just because an intermediate pull is empty/unchanged. Use `progress_validated` vs `candidate_records` to track whether more results may still appear (`progress_validated < candidate_records`). If transport/session fails, resume using the same `job_id`.

Retrieve plan features and current usage limits for your API key. Use when: - You want to know how many records/jobs/monitors your plan allows. - You want to check current usage against plan limits before running a large job.

List all jobs submitted by you. Returns your job history with IDs, queries, statuses, and timestamps.

Append entities from a CSV file to an existing dataset. Parses the CSV and appends its entities to the dataset. Each row must have a `name` column; include a `domain` or `description` column (or both) for meaningful enrichment. Duplicate rows (by name) are skipped. To create a new dataset from a CSV, use `create_dataset_from_csv` instead.

Map a resource (job, monitor, or monitor_group) to a webhook. Use when: - You want a webhook to fire for a specific job or monitor's deliveries.

Expand a job by processing more records beyond the initial limit. This increases the number of records the system processes (which costs additional credits). Only use this when the user wants MORE data processed. This only applies to jobs originally submitted with `limit`. If a job was submitted without `limit`, there is nothing to continue. The new_limit must be greater than the previous limit when provided. If omitted, API defaults to your plan maximum.

List your datasets.

Disable a monitor to stop its scheduled runs. The monitor can be re-enabled later with enable_monitor.

Enable a previously disabled monitor to resume its scheduled runs.

Check API health status. This tool maps to GET /health and does not require an API key.

Preview suggested validators, enrichments, and date ranges before submitting. Use when: - You want to inspect/edit auto-generated validators/enrichments before submitting. - You want to preview date adjustments via `date_modification_message`. Do not use when: - You want to start processing immediately with final inputs (use `submit_query`). Key behavior: - Preview-only endpoint: does not create a job and does not start processing. - Suggestions are LLM-generated and not deterministic across calls. - To reuse suggestions, pass them explicitly to `submit_query`.

List all your monitors. Returns all monitors with their schedule, status, reference query, and webhook config.

List your projects.

List the webhooks mapped to a specific resource (job/monitor/monitor_group). Use when: - You have a job or monitor ID and want to know which webhooks will fire for it.

List source groups (named domain allowlists). Use when: - You want to discover reusable, named sets of source domains (public groups plus any organization-visibility groups your organization can access). - You need a group's `slug` to attach to `submit_query` via `source_groups` to scope fetching to that domain allowlist.

Create a new CatchAll processing job from a natural-language query. Use when: - You want to start a new CatchAll web research run from a user query. - You want the API to fetch/process sources and then return structured results. Do not use when: - You want status for an existing job (use `get_job_status`). - You want records for an existing job (use `pull_results`). Key rules: - `query` is required. - You can submit with only `query`; omitted optional fields (`validators`, `enrichments`, `start_date`, `end_date`) are auto-selected/generated by the API. - Optional fields are independent: you can pass any subset (for example, custom `validators` but no `enrichments`), and omitted fields are still auto-selected/generated. - When `connected_dataset_ids` is set, the `query` must describe the **topic or event type only** (e.g. "M&A activity", "regulatory filings", "executive changes"). Do NOT write things like "for my companies", "for the selected list of companies", or "news about my watchlist" — the entity filtering is applied automatically by the connected dataset. Mentioning companies in the query when a dataset is attached is redundant and degrades retrieval quality. - When `connected_dataset_ids` is set, entity-relevance validators (e.g. `company_is_primary_subject`) are generated automatically by the API. Do NOT add them manually to `validators` — they are redundant and may conflict with the auto-generated ones. Only pass validators that describe the **event or topic**, not entity filtering. - `start_date` and `end_date` filter by web page discovery date, not event date. - Discovery dates and extracted event dates can differ. For event-time accuracy, use event-focused validators/enrichments and verify `event_date` in pulled results. - `end_date` must be after `start_date`. - Dates outside your plan lookback limits return API 400. - `limit` controls processed record count (cost-affecting). Omit it to retrieve everything up to your plan's maximum. If provided, must be >= 10. - `validators` / `enrichments` may be passed either as arrays or as JSON-string arrays (for client compatibility). - `validators[].type` must be `boolean` (if omitted, it defaults to `boolean`). - `enrichments[].type` supported values: text, number, date, option, url, company. Basic examples: - validators: `[{"name":"is_acquisition_event","description":"true if page describes an acquisition","type":"boolean"}]` - enrichments: `[{"name":"acquiring_company","description":"Extract acquiring company","type":"company"},{"name":"deal_value","description":"Extract announced deal value","type":"number"}]` Next step: - Save the returned `job_id`. - Poll `get_job_status` and call `pull_results` (partial results can appear before completion).

Send a test delivery to a webhook endpoint. Use when: - You want to verify a webhook URL is reachable and correctly configured before attaching it to a monitor or job.

Manually trigger webhook delivery for a resource (job/monitor/monitor_group). Use when: - You want to (re-)send a webhook delivery on demand instead of waiting for the automatic dispatch — e.g. to replay a missed or failed delivery.

Check the quality of a query before submitting a job ("Check Query Quality"). Use when: - You want quick feedback on whether a query is well-formed for CatchAll before spending credits on a job. - You want concrete suggestions to improve a vague or overly broad query. Do not use when: - You want to preview auto-generated validators/enrichments (use `initialize_query`). - You want to actually run a search (use `submit_query`).

Get current API version. This tool maps to GET /version and does not require an API key.

List all your webhooks. Use when: - You want to see all webhook endpoints configured in your account. - You need to find a webhook_id to pass to monitors (via webhook_ids) or jobs.