Mostly Right changelog

• + ask_support
• + get_support_answer

https://mostlyright.md/llms.txt

− Building is free with an account: datasets, recipes, runs, refreshes, the MCP build tools and mr_live_ keys. The first 500 rows of every table, in stored order, are free to everyone, with or without an account. Without an account you get that sample and nothing else: no filter, sort, search, download or page past row 500. A free workspace may read 100,000 rows and run 1,000 queries a UTC month past the sample. The allowance is shared by everyone in the workspace, and a free workspace may have an− The same server also builds datasets, free with an account, over an OAuth connection carrying the datasets:build scope: create_dataset, update_dataset, configure_table_view, generate_cover, set_dataset_visibility, archive_dataset, list_my_datasets, get_my_dataset, list_dataset_events, list_recipes, get_recipe, register_recipe, explain_recipe, forecast_recipe, test_recipe, propose_revision, replay_run, start_run, confirm_run, approve_full_run, cancel_run, get_run, run_events, list_runs, compare_r+ There are three tiers. Reading metadata, schemas and sample rows needs no account, and neither does a bounded query of up to 100 rows a page (POST /api/public/tables/{table_id}/query). Querying a whole table — up to 10000 rows a page, paged to the end — and downloading a full snapshot need a free account and an API key from Settings → Access; the first request connects the dataset to your workspace. Building a new dataset — writing a recipe, running it and publishing the result — needs a paid su+ The same server also builds datasets, on a paid workspace, over an OAuth connection carrying the datasets:build scope: create_dataset, update_dataset, configure_dataset_layout, configure_table_view, generate_cover, set_dataset_visibility, archive_dataset, list_my_datasets, get_my_dataset, list_dataset_events, list_recipes, get_recipe, register_recipe, explain_recipe, forecast_recipe, test_recipe, propose_revision, replay_run, start_run, confirm_run, approve_full_run, cancel_run, get_run, run_eve

• + configure_dataset_layout
• + configure_table_view
• + list_models
• + suggest_enrichments

https://mostlyright.md/llms.txt

− There are three tiers. Reading metadata, schemas and sample rows needs no account, and neither does a bounded query of up to 100 rows a page (POST /api/public/tables/{table_id}/query). Querying a whole table — up to 10000 rows a page, paged to the end — and downloading a full snapshot need a free account and an API key from Settings → Access; the first request connects the dataset to your workspace. Building a new dataset — writing a recipe, running it and publishing the result — needs a paid su− Agents can use the MCP server at /api/mcp instead: Streamable HTTP, protocol 2025-11-25 (2025-06-18, 2025-03-26 and 2024-11-05 also negotiate), stateless, no session id. Server card: /.well-known/mcp.json. search, fetch, search_datasets, get_dataset, list_tables, get_table_schema, sample_rows, get_download_instructions and get_access_instructions answer with NO credential; search and fetch are the two-tool connector contract, where search returns {id, title, url} per dataset and fetch returns th− The same server also builds datasets, on a paid workspace, over an OAuth connection carrying the datasets:build scope: create_dataset, update_dataset, generate_cover, set_dataset_visibility, archive_dataset, list_my_datasets, get_my_dataset, list_dataset_events, list_recipes, get_recipe, register_recipe, explain_recipe, forecast_recipe, test_recipe, propose_revision, replay_run, start_run, confirm_run, approve_full_run, cancel_run, get_run, run_events, list_runs, compare_runs, query_run, run_art+ Building is free with an account: datasets, recipes, runs, refreshes, the MCP build tools and mr_live_ keys. The first 500 rows of every table, in stored order, are free to everyone, with or without an account. Without an account you get that sample and nothing else: no filter, sort, search, download or page past row 500. A free workspace may read 100,000 rows and run 1,000 queries a UTC month past the sample. The allowance is shared by everyone in the workspace, and a free workspace may have an+ - Documentation index (Markdown): https://mostlyright.md/docs/llms.txt+ - Full documentation (Markdown): https://mostlyright.md/docs/llms-full.txt+ - Search API: https://mostlyright.md/docs/reference/search/+ - JavaScript SDK and native fetch: https://mostlyright.md/docs/guides/javascript/+ Agents can use the MCP server at /api/mcp instead: Streamable HTTP, protocol 2025-11-25 (2025-06-18, 2025-03-26 and 2024-11-05 also negotiate), stateless, no session id. Server card: /.well-known/mcp.json. search, fetch, search_datasets, get_dataset, list_tables, get_table_schema, sample_rows, get_download_instructions and get_access_instructions answer with NO credential; search and fetch are the two-tool connector contract, where search returns {id, title, url} per dataset and fetch returns th+ The same server also builds datasets, free with an account, over an OAuth connection carrying the datasets:build scope: create_dataset, update_dataset, configure_table_view, generate_cover, set_dataset_visibility, archive_dataset, list_my_datasets, get_my_dataset, list_dataset_events, list_recipes, get_recipe, register_recipe, explain_recipe, forecast_recipe, test_recipe, propose_revision, replay_run, start_run, confirm_run, approve_full_run, cancel_run, get_run, run_events, list_runs, compare_r

• + search_table
• + generate_cover
• + set_dataset_visibility
• + archive_dataset
• + list_dataset_events
• + list_recipes
• + get_recipe
• + compare_runs
• + sample_table
• + list_column_values
• + preview_table_search
• + list_notes
• + list_table_parts
• + verify_run
• + reschedule_table
• + demote_table
• + refresh_table
• + resync_table
• + archive_table
• + rename_table
• + probe_source
• + get_source_probe
• + get_run_attempts
• + get_recipe_readiness
• + inspect_table_search
• + configure_table_search
• + build_table_search
• + activate_table_search
• + get_search_job
• + get_search_evaluation
• + add_search_cases
• + cancel_search_job
• + register_stream_document
• + create_stream_registry
• + start_stream
• + get_stream
• + control_stream_subscription

https://mostlyright.md/llms.txt

− Mostly Right publishes open, machine-readable datasets as immutable Apache Parquet versions: time series, benchmarks and reference data for analysis and machine learning. Every dataset carries a documented column schema, column profiles, an explicit licence and a named publisher.− The same server also builds datasets, on a paid workspace, over an OAuth connection carrying the datasets:build scope: create_dataset, update_dataset, list_my_datasets, get_my_dataset, register_recipe, explain_recipe, forecast_recipe, test_recipe, propose_revision, replay_run, start_run, confirm_run, approve_full_run, cancel_run, get_run, run_events, list_runs, query_run, run_artifacts, get_artifact_download, write_note, promote_table, get_table, diagnose_table, list_source_credentials, normaliz− - https://mostlyright.md/docs/recipes/market-settlement-join/ — settlement rules from two prediction venues joined on one event.+ Mostly Right is where you publish and discover datasets with AI agents. You describe a dataset. Agents find the sources and configure the ingest engine. Mostly Right runs the build, keeps the dataset updated, and serves each version as Apache Parquet or as JSON through the API and MCP.+ The catalog covers AI work: specialized and frontier models, data ETL, agentic engineering, and the records those workflows use. Examples include LLM benchmark scores by model since 2019, Y Combinator company status and Form D filings, Czech companies with KYB risk flags, EU financial firm licences and regulator warnings, and movies and TV shows with plots, cast and ratings. Every dataset carries a documented column schema, column profiles, an explicit licence and a named publisher.+ The same server also builds datasets, on a paid workspace, over an OAuth connection carrying the datasets:build scope: create_dataset, update_dataset, generate_cover, set_dataset_visibility, archive_dataset, list_my_datasets, get_my_dataset, list_dataset_events, list_recipes, get_recipe, register_recipe, explain_recipe, forecast_recipe, test_recipe, propose_revision, replay_run, start_run, confirm_run, approve_full_run, cancel_run, get_run, run_events, list_runs, compare_runs, query_run, run_art+ - https://mostlyright.md/docs/recipes/benchmark-scores-join/ — LLM benchmark scores joined to model release dates.

• + Recipe patterns by source shape

• + explain_recipe
• + forecast_recipe
• + test_recipe

https://mostlyright.md/llms.txt

− The same server also builds datasets, on a paid workspace, over an OAuth connection carrying the datasets:build scope: create_dataset, update_dataset, list_my_datasets, get_my_dataset, register_recipe, propose_revision, replay_run, start_run, confirm_run, approve_full_run, cancel_run, get_run, run_events, list_runs, query_run, run_artifacts, get_artifact_download, write_note, promote_table, get_table, diagnose_table, list_source_credentials, normalize_reader_options, start_source_inspection and + The same server also builds datasets, on a paid workspace, over an OAuth connection carrying the datasets:build scope: create_dataset, update_dataset, list_my_datasets, get_my_dataset, register_recipe, explain_recipe, forecast_recipe, test_recipe, propose_revision, replay_run, start_run, confirm_run, approve_full_run, cancel_run, get_run, run_events, list_runs, query_run, run_artifacts, get_artifact_download, write_note, promote_table, get_table, diagnose_table, list_source_credentials, normaliz