此 MCP 可以做什么
Lets users explore and query a data warehouse through an AI data analyst agent.
功能
工具
19
工具
myriade_ask
Ask a natural language question about a database using Myriade's
data analyst agent.
The agent will explore the database schema, search the catalog, run SQL
queries, and analyze the results to answer your question. Use this for
complex analytical questions that require multiple steps; for simple
schema lookups or direct SQL, prefer the other tools.
The run executes in the background as a regular Myriade conversation
(visible in the web UI). If it finishes quickly the answer is returned
directly; otherwise the tool returns ``{"status": "running",
"conversation_id": ...}`` — poll myriade_get_answer with that id to
retrieve the result. Long analytical questions are expected to take
several minutes.
Note: the agent has access to catalog write operations (create/update
assets, tags, terms) — it may modify catalog state to record findings.
It can also read, create and edit reports (ask it to "write a report
on ..." or "update report <id> with ..."). Use myriade_query directly
for strictly read-only inspection.
Requires AI_PROVIDER and the corresponding API key (e.g.
AI_PROVIDER=anthropic + ANTHROPIC_API_KEY).
可能执行破坏性操作
输入模式
{'type': 'object', 'title': 'askArguments', 'required': ['database_id', 'question'], 'properties': {'question': {'type': 'string', 'title': 'Question', 'maxLength': 4000, 'minLength': 1, 'description': 'Natural-language question about the data'}, 'database_id': {'type': 'string', 'title': 'Database Id', 'format': 'uuid', 'description': 'UUID of the database'}}}
输出模式
{'type': 'object', 'title': 'askOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
myriade_get_answer
Retrieve the result of a myriade_ask run.
Returns ``{"status": "completed", "answer": ...}`` once the analyst
agent has answered, or ``{"status": "running"}`` while it is still
working — keep polling in that case. ``{"status": "failed"}`` /
``{"status": "quota_exceeded"}`` carry the run's error. ``{"status":
"no_answer"}`` means the run is gone without a recorded answer (agent
stopped silently, or the server restarted): start a new myriade_ask run.
只读
幂等
输入模式
{'type': 'object', 'title': 'get_answerArguments', 'required': ['conversation_id'], 'properties': {'wait_seconds': {'type': 'integer', 'title': 'Wait Seconds', 'default': 20, 'maximum': 30, 'minimum': 0, 'description': 'How long to wait for the run to finish before reporting its status (long-poll)'}, 'conversation_id': {'type': 'string', 'title': 'Conversation Id', 'format': 'uuid', 'description': 'Conversation UUID returned by myriade_ask'}}}
输出模式
{'type': 'object', 'title': 'get_answerOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
myriade_get_check
Get a monitoring check (name, SQL, severity, cadence, status) and its
most recent run results, newest first.
On governed databases the SQL and evidence rows are redacted unless you
manage the database or authored the proposal.
只读
幂等
输入模式
{'type': 'object', 'title': 'get_checkArguments', 'required': ['check_id'], 'properties': {'limit': {'type': 'integer', 'title': 'Limit', 'default': 20, 'maximum': 100, 'minimum': 1, 'description': 'Run results to return (1-100)'}, 'check_id': {'type': 'string', 'title': 'Check Id', 'format': 'uuid', 'description': 'UUID of the monitoring check'}}}
输出模式
{'type': 'object', 'title': 'get_checkOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
myriade_get_database_info
Get detailed information about a specific database connection.
Returns information about the database including its name, engine type,
description, write mode, and any AI memory notes.
只读
幂等
输入模式
{'type': 'object', 'title': 'get_database_infoArguments', 'required': ['database_id'], 'properties': {'database_id': {'type': 'string', 'title': 'Database Id', 'format': 'uuid', 'description': 'UUID of the database (from myriade_list_databases)'}}}
输出模式
{'type': 'object', 'title': 'get_database_infoOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
myriade_get_query_results
Re-execute a saved Myriade query and return fresh results.
The query's SQL always re-runs under the caller's governance policies
(table permissions, column masking) — results are never served from
Myriade's cached snapshot.
Prefer this over myriade_query for recurring metrics (dashboards,
cockpits): the SQL stays governed in Myriade instead of travelling
through prompts. Write statements are blocked. Returns up to 50 rows
with column metadata, same shape as myriade_query.
只读
幂等
输入模式
{'type': 'object', 'title': 'get_query_resultsArguments', 'required': ['query_id'], 'properties': {'query_id': {'type': 'string', 'title': 'Query Id', 'format': 'uuid', 'description': 'UUID of a saved Myriade query (from a Myriade query link, e.g. in myriade_ask responses)'}}}
输出模式
{'type': 'object', 'title': 'get_query_resultsOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
myriade_get_table_schema
Get the column-level schema for a specific table.
Returns column names, data types, and descriptions for the table,
plus the SQL `dialect` to use with myriade_query on this database.
When the name is ambiguous across schemas, the candidates are returned
under `matches` instead — pass `schema_name` to pick one.
Raises if the user has no access to the table under governance.
只读
幂等
输入模式
{'type': 'object', 'title': 'get_table_schemaArguments', 'required': ['database_id', 'table_name'], 'properties': {'table_name': {'type': 'string', 'title': 'Table Name', 'maxLength': 255, 'minLength': 1, 'description': 'Name of the table to inspect'}, 'database_id': {'type': 'string', 'title': 'Database Id', 'format': 'uuid', 'description': 'UUID of the database'}, 'schema_name': {'anyOf': [{'type': 'string', 'maxLength': 255}, {'type': 'null'}], 'title': 'Schema Name', 'default': None, 'description': 'Optional schema name to disambiguate same-named tables'}}}
输出模式
{'type': 'object', 'title': 'get_table_schemaOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
myriade_list_databases
List databases configured in Myriade (paginated, sorted by name).
Returns paginated database connections with their IDs, names, engines,
and descriptions. Use the database ID with other tools.
只读
幂等
输入模式
{'type': 'object', 'title': 'list_databasesArguments', 'properties': {'limit': {'type': 'integer', 'title': 'Limit', 'default': 50, 'maximum': 200, 'minimum': 1, 'description': 'Maximum results to return (1-200)'}, 'offset': {'type': 'integer', 'title': 'Offset', 'default': 0, 'minimum': 0, 'description': 'Number of results to skip for pagination'}}}
输出模式
{'type': 'object', 'title': 'list_databasesOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
myriade_list_reports
List or search reports (documents) written in Myriade.
Without ``query`` returns the most recently updated reports first; with
``query`` returns the best matches first. Only reports the caller may see
are returned. Read one with myriade_read_report.
只读
幂等
输入模式
{'type': 'object', 'title': 'list_reportsArguments', 'properties': {'limit': {'type': 'integer', 'title': 'Limit', 'default': 20, 'maximum': 50, 'minimum': 1, 'description': 'Maximum results to return (1-50)'}, 'query': {'anyOf': [{'type': 'string', 'maxLength': 200, 'minLength': 1}, {'type': 'null'}], 'title': 'Query', 'default': None, 'description': 'Optional search terms (title, description, content); results are ranked by relevance'}, 'database_id': {'anyOf': [{'type': 'string', 'format': 'uuid'}, {'type': 'null'}], 'title': 'Database Id', 'default': None, 'description': 'UUID of the database whose reports to list. Omit for organisation-level reports (not bound to a database).'}}}
输出模式
{'type': 'object', 'title': 'list_reportsOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
myriade_list_tables
List tables in a database (paginated, sorted by database/schema/name).
Returns a flat, paginated list of tables. Each item has database_name,
schema, and name fields. The response's `dialect` is the SQL dialect
to use with myriade_query on this database.
只读
幂等
输入模式
{'type': 'object', 'title': 'list_tablesArguments', 'required': ['database_id'], 'properties': {'limit': {'type': 'integer', 'title': 'Limit', 'default': 100, 'maximum': 500, 'minimum': 1, 'description': 'Maximum tables to return (1-500)'}, 'offset': {'type': 'integer', 'title': 'Offset', 'default': 0, 'minimum': 0, 'description': 'Number of tables to skip for pagination'}, 'database_id': {'type': 'string', 'title': 'Database Id', 'format': 'uuid', 'description': 'UUID of the database'}}}
输出模式
{'type': 'object', 'title': 'list_tablesOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
myriade_monitoring_overview
Snapshot of a database's monitoring board: check count, standing alerts
(open / acknowledged / known) with their headlines, recent notifications
and triage notes, followed by the open alerts as JSON — each with its
investigation and evidence while they fit, then as listing lines, with
``shown`` / ``total`` / ``detailed`` saying which.
Alert ids feed myriade_triage_alert; check ids feed myriade_get_check.
Investigation details are redacted unless you manage the database.
只读
幂等
输入模式
{'type': 'object', 'title': 'monitoring_overviewArguments', 'required': ['database_id'], 'properties': {'database_id': {'type': 'string', 'title': 'Database Id', 'format': 'uuid', 'description': 'UUID of the database'}}}
输出模式
{'type': 'object', 'title': 'monitoring_overviewOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
myriade_propose_check
Propose a monitoring check for the user to review in Myriade.
This creates a PROPOSAL, never an active check: nothing runs until a
database owner or org admin accepts it in the app. The SQL is test-run
once under your governance context to validate it.
Contract for ``check_sql``: it MUST return rows ONLY when something is
wrong (zero rows = healthy) and use relative time (``now() - interval``),
never absolute dates. Minimum cadence is 5 minutes. Names are unique per
database — read myriade_monitoring_overview before proposing a
near-duplicate.
输入模式
{'type': 'object', 'title': 'propose_checkArguments', 'required': ['database_id', 'name', 'check_sql'], 'properties': {'name': {'type': 'string', 'title': 'Name', 'maxLength': 255, 'minLength': 1, 'description': 'Short human-readable name (unique per database)'}, 'severity': {'enum': ['LOW', 'MEDIUM', 'HIGH', 'CRITICAL'], 'type': 'string', 'title': 'Severity', 'default': 'MEDIUM', 'description': 'Severity raised when the check trips'}, 'check_sql': {'type': 'string', 'title': 'Check Sql', 'maxLength': 20000, 'minLength': 1, 'description': 'SQL returning zero rows when healthy, rows when wrong'}, 'group_name': {'anyOf': [{'type': 'string', 'maxLength': 255}, {'type': 'null'}], 'title': 'Group Name', 'default': None, 'description': "Cluster to file the check under (reuse an existing group's exact name), or omit to leave it unfiled"}, 'database_id': {'type': 'string', 'title': 'Database Id', 'format': 'uuid', 'description': 'UUID of the database'}, 'description': {'type': 'string', 'title': 'Description', 'default': '', 'maxLength': 2000, 'description': 'One sentence, prospective tense: "Trips when <condition> (baseline ~X)"'}, 'interval_minutes': {'type': 'integer', 'title': 'Interval Minutes', 'default': 60, 'minimum': 5, 'description': 'Cadence in minutes (minimum 5)'}}}
输出模式
{'type': 'object', 'title': 'propose_checkOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
myriade_query
Execute a read-only SQL query against a database.
Write the SQL in the dialect of the target database's engine — see the
`dialect` field returned by myriade_list_tables and
myriade_get_table_schema, or the engine in myriade_list_databases.
Write operations (INSERT, UPDATE, DELETE, DROP, etc.) are blocked.
Returns up to 50 rows of results with column metadata. Successful
queries are saved in Myriade; the response's `query_id` can be
replayed later with myriade_get_query_results without resending SQL.
只读
幂等
输入模式
{'type': 'object', 'title': 'queryArguments', 'required': ['database_id', 'sql'], 'properties': {'sql': {'type': 'string', 'title': 'Sql', 'maxLength': 20000, 'minLength': 1, 'description': "Read-only SQL query, written for the target database's dialect (writes are blocked)"}, 'title': {'anyOf': [{'type': 'string', 'maxLength': 255, 'minLength': 1}, {'type': 'null'}], 'title': 'Title', 'default': None, 'description': 'Optional name for the saved query, shown in Myriade'}, 'database_id': {'type': 'string', 'title': 'Database Id', 'format': 'uuid', 'description': 'UUID of the database'}}}
输出模式
{'type': 'object', 'title': 'queryOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
myriade_read_asset
Get detailed information about a catalog object (table, column, schema,
or glossary term — assets and terms share one id space).
Returns the asset's description, tags, status, columns (for tables),
sample data, parent hierarchy, and activity feed; for a glossary term,
its definition, synonyms, related terms, and activity feed.
只读
幂等
输入模式
{'type': 'object', 'title': 'read_assetArguments', 'required': ['database_id', 'asset_id'], 'properties': {'asset_id': {'type': 'string', 'title': 'Asset Id', 'format': 'uuid', 'description': 'UUID of the catalog object (asset or glossary term)'}, 'database_id': {'type': 'string', 'title': 'Database Id', 'format': 'uuid', 'description': 'UUID of the database'}}}
输出模式
{'type': 'object', 'title': 'read_assetOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
myriade_read_report
Read a report's full markdown content.
Applies the same visibility rule as the web app; a report the caller may
not see returns a "not found" error. To create or edit a report, use
myriade_ask — the analyst agent knows how reports are structured.
只读
幂等
输入模式
{'type': 'object', 'title': 'read_reportArguments', 'required': ['document_id'], 'properties': {'document_id': {'type': 'string', 'title': 'Document Id', 'format': 'uuid', 'description': 'UUID of the report (from myriade_list_reports)'}}}
输出模式
{'type': 'object', 'title': 'read_reportOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
myriade_request_catalog_change
Propose a catalog change for the data team to review (no edit rights needed).
Creates a reviewable suggestion routed to the asset's reviewers. Does NOT
change the published catalog until approved.
Use this instead of ``myriade_update_asset`` when you lack edit rights or
are not confident enough to write directly.
输入模式
{'type': 'object', 'title': 'request_catalog_changeArguments', 'required': ['database_id', 'asset_id', 'field', 'proposed_value', 'rationale'], 'properties': {'field': {'enum': ['description', 'tags'], 'type': 'string', 'title': 'Field', 'description': 'Field to change: "description" or "tags"'}, 'asset_id': {'type': 'string', 'title': 'Asset Id', 'format': 'uuid', 'description': 'UUID of the asset to propose a change for'}, 'rationale': {'type': 'string', 'title': 'Rationale', 'maxLength': 2000, 'description': 'Why this change is proposed — shown to reviewers and in the activity feed'}, 'database_id': {'type': 'string', 'title': 'Database Id', 'format': 'uuid', 'description': 'UUID of the database'}, 'proposed_value': {'type': 'string', 'title': 'Proposed Value', 'maxLength': 10000, 'description': 'Proposed description text, or JSON-encoded list of tag names'}}}
输出模式
{'type': 'object', 'title': 'request_catalog_changeOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
myriade_sample_data
Get sample rows from a table to understand the data.
Returns a random sample of up to ``limit`` rows (max 20).
只读
输入模式
{'type': 'object', 'title': 'sample_dataArguments', 'required': ['database_id', 'table_name', 'schema_name'], 'properties': {'limit': {'type': 'integer', 'title': 'Limit', 'default': 10, 'maximum': 20, 'minimum': 1, 'description': 'Rows to sample (1-20)'}, 'table_name': {'type': 'string', 'title': 'Table Name', 'maxLength': 255, 'minLength': 1, 'description': 'Name of the table to sample'}, 'database_id': {'type': 'string', 'title': 'Database Id', 'format': 'uuid', 'description': 'UUID of the database'}, 'schema_name': {'type': 'string', 'title': 'Schema Name', 'maxLength': 255, 'minLength': 1, 'description': 'Schema containing the table'}, 'database_name': {'anyOf': [{'type': 'string', 'maxLength': 255}, {'type': 'null'}], 'title': 'Database Name', 'default': None, 'description': 'Optional warehouse-side database name (e.g. for Snowflake)'}}}
输出模式
{'type': 'object', 'title': 'sample_dataOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
myriade_search_catalog
Search the data catalog for tables, columns, or terms by name/description.
Uses fuzzy matching and full-text search. Keep queries short
(1-2 words) for best results.
只读
幂等
输入模式
{'type': 'object', 'title': 'search_catalogArguments', 'required': ['database_id', 'text'], 'properties': {'text': {'type': 'string', 'title': 'Text', 'maxLength': 200, 'minLength': 1, 'description': 'Search query — keep short (1-2 words) for best results'}, 'limit': {'type': 'integer', 'title': 'Limit', 'default': 20, 'maximum': 50, 'minimum': 1, 'description': 'Maximum results to return (1-50)'}, 'asset_type': {'anyOf': [{'enum': ['TABLE', 'COLUMN', 'TERM'], 'type': 'string'}, {'type': 'null'}], 'title': 'Asset Type', 'default': None, 'description': 'Filter: "TABLE", "COLUMN", or "TERM"'}, 'database_id': {'type': 'string', 'title': 'Database Id', 'format': 'uuid', 'description': 'UUID of the database'}}}
输出模式
{'type': 'object', 'title': 'search_catalogOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
myriade_triage_alert
Triage an alert (requires manage access: database owner or org admin).
- ``resolve``: the underlying problem is handled or gone.
- ``acknowledge``: keep it standing with a ``statement`` of what is
happening — only for a situation that may worsen.
- ``dismiss``: false positive / expected; refused while the check still
trips (it would just respawn).
- ``snooze``: silence for ``hours`` with a ``statement``; reopens by itself
if the check still trips afterwards.
- ``reopen``: send a parked alert back to "needs action" with a
``statement`` of what is no longer covered.
可能执行破坏性操作
输入模式
{'type': 'object', 'title': 'triage_alertArguments', 'required': ['alert_id', 'action'], 'properties': {'note': {'anyOf': [{'type': 'string', 'maxLength': 2000}, {'type': 'null'}], 'title': 'Note', 'default': None, 'description': 'Optional note appended to the alert details (resolve, dismiss)'}, 'hours': {'anyOf': [{'type': 'integer', 'maximum': 168, 'minimum': 1}, {'type': 'null'}], 'title': 'Hours', 'default': None, 'description': 'Snooze window in hours (1-168). Required for snooze.'}, 'action': {'enum': ['resolve', 'acknowledge', 'dismiss', 'snooze', 'reopen'], 'type': 'string', 'title': 'Action', 'description': 'Transition to apply to the alert'}, 'alert_id': {'type': 'string', 'title': 'Alert Id', 'format': 'uuid', 'description': 'UUID of the alert'}, 'statement': {'anyOf': [{'type': 'string', 'maxLength': 500}, {'type': 'null'}], 'title': 'Statement', 'default': None, 'description': 'One SHORT present-tense plain-text sentence (~120 chars, no Markdown) shown as the alert headline. Required for acknowledge, snooze and reopen.'}}}
输出模式
{'type': 'object', 'title': 'triage_alertOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
myriade_update_asset
Update a catalog asset's documentation (requires edit rights).
- ``description`` replaces the existing description (auto-sets status to
"draft" if status is null).
- ``tag_ids`` replaces all tags (accepts names or UUIDs; auto-creates
missing tags).
- ``status`` is "draft" (needs review) or "published". Only use
"published" with high confidence.
Use ``myriade_request_catalog_change`` instead when you lack edit rights
or are not confident enough to write directly.
可能执行破坏性操作
输入模式
{'type': 'object', 'title': 'update_assetArguments', 'required': ['database_id', 'asset_id'], 'properties': {'status': {'anyOf': [{'enum': ['draft', 'published'], 'type': 'string'}, {'type': 'null'}], 'title': 'Status', 'default': None, 'description': '"draft" (needs review) or "published" (production-ready)'}, 'tag_ids': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}, 'maxItems': 50}, {'type': 'null'}], 'title': 'Tag Ids', 'default': None, 'description': 'Tag names or UUIDs (replaces all tags). Auto-creates missing tags.'}, 'asset_id': {'type': 'string', 'title': 'Asset Id', 'format': 'uuid', 'description': 'UUID of the asset to update'}, 'database_id': {'type': 'string', 'title': 'Database Id', 'format': 'uuid', 'description': 'UUID of the database'}, 'description': {'anyOf': [{'type': 'string', 'maxLength': 10000}, {'type': 'null'}], 'title': 'Description', 'default': None, 'description': "Replaces existing description; auto-sets status='draft' if status is null"}}}
输出模式
{'type': 'object', 'title': 'update_assetOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
历史
近期工具变更
已添加
myriade_triage_alert
2026年9月17日 07:57
已添加
myriade_propose_check
2026年9月17日 07:57
已添加
myriade_get_check
2026年9月17日 07:57
已添加
myriade_monitoring_overview
2026年9月17日 07:57
已添加
myriade_get_answer
2026年9月17日 07:57
已添加
myriade_ask
2026年9月17日 07:57
已添加
myriade_read_report
2026年9月17日 07:57
已添加
myriade_list_reports
2026年9月17日 07:57
已添加
myriade_request_catalog_change
2026年9月17日 07:57
已添加
myriade_update_asset
2026年9月17日 07:57
已添加
myriade_read_asset
2026年9月17日 07:57
已添加
myriade_search_catalog
2026年9月17日 07:57
已添加
myriade_sample_data
2026年9月17日 07:57
已添加
myriade_get_query_results
2026年9月17日 07:57
已添加
myriade_query
2026年9月17日 07:57
已添加
myriade_get_table_schema
2026年9月17日 07:57
已添加
myriade_list_tables
2026年9月17日 07:57
已添加
myriade_get_database_info
2026年9月17日 07:57
已添加
myriade_list_databases
2026年9月17日 07:57
发现
类似的 MCP 服务器
Huggingface
io.github.pipeworx-io/huggingface
Provides access to Hugging Face model, dataset, and Space metadata, alongside broader structured research and data-routing tools.
Ai Model Experiments
io.github.pipeworx-io/ai-model-experiments
Runs prompts across multiple AI models and compares their outputs, costs, latency, token usage, and errors.
branchly
io.branchly/branchly
Manages an AI application’s knowledge-base content, prompts, tools, data sources, retrieval context, sessions, and usage analytic…
HuggingFace New Dataset Release Tracker (hfdatasets)
com.a2awire/data-new-dataset-release-huggingface-tracker
Tracks newly released Hugging Face datasets, including updates, tags, and licenses.
aifu Agent Market
asia.aifu/agent-market
Discovers and hires agents or approved humans through a job ledger, verifies agent performance, and provides crypto, equity, sect…
SigRank — AI Operator Benchmarking
com.signalaf/sigrank
Benchmarks AI operator token usage, calculates cascade metrics, compares leaderboard performance, diagnoses efficiency, and simul…
Automan
io.github.i02202/automan
Offers paid AI skills for data analysis, classification, extraction, reporting, summarization, translation, text editing, copywri…
Nonobench
io.github.mauricekleine/nonobench
Provides nonogram puzzles and benchmark data for comparing language-model accuracy, cost, latency, token use, and solution outcom…