> ## Documentation Index
> Fetch the complete documentation index at: https://platform.atlan.com/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> To act on Atlan objects, use the Atlan MCP server at https://api.atlan.com/mcp or the atlanai CLI; `atlanai --map json` prints its command map. Run a read-only identity check before any write.
> The docs MCP server at /mcp searches these docs only. It cannot read or change Atlan objects.
> SDK packages: Python `atlanai` (PyPI) and TypeScript `@atlanai/sdk` (npm). Show Python first, then TypeScript.

# Understand the tool catalog

> Find every first-party MCP tool and distinguish it from account-specific downstream tools.

Atlan MCP returns one catalog assembled from Registry, installed extensions,
and registered downstream MCP servers. This page records the first-party tools
declared by Agent Gateway at commit `76911562`.

Always inspect the authenticated MCP `tools/list` result before calling a tool.
It is the authority for the deployment, account, exact wire name, and current
input schema you are connected to.

## Provider summary

| Provider | Tools | Current scope |
| - | -: | - |
| `registry` | 16 | Artifact discovery, reads, writes, relationships, account traces, and session initialization |
| `agent` | 6 | Bulk Agent creation plus Agent, harness, and session trace reads |
| `skill` | 5 | Skill creation from uploaded bundles and Skill trace reads |
| `file` | 1 | Reserve an upload destination |
| `ontology` | 4 | Retrieve bounded release context, export or preview interchange documents, and plan SPARQL metadata queries |
| `api` | 0 | No extension-owned tools |
| `eval` | 0 | No extension-owned tools |
| `mcp` | 0 | Aggregates and routes tools but declares none of its own |
| `model` | 0 | No extension-owned tools |
| `otel` | 0 | No extension-owned tools |
| `secret` | 0 | No extension-owned tools |

## Registry tools

| Wire name | Effect | Use |
| - | - | - |
| `registry__initialise` | Write | Open an auditable MCP session. |
| `registry__search_artifacts` | Read | Search visible artifacts by text, kind, or semantic similarity. |
| `registry__get_artifact` | Read | Read one artifact by handle. |
| `registry__list_artifacts` | Read | Browse the latest artifacts of one kind. |
| `registry__list_workspaces` | Read | List workspaces available to the caller. |
| `registry__get_related_artifacts` | Read | Read typed relationships in both directions. |
| `registry__get_kind_schema` | Read | Read a kind's fields and allowed relationship types. |
| `registry__create_artifact` | Write | Create one JSON-backed artifact. |
| `registry__create_artifacts` | Write | Create up to 100 JSON-backed artifacts of one kind. |
| `registry__patch_artifact` | Write | Change selected artifact fields. |
| `registry__archive_artifact` | Destructive write | Archive an artifact while retaining its content. |
| `registry__relate_artifact` | Write | Create or update a typed relationship. |
| `registry__create_artifact_linked` | Write | Create an artifact and relate it to an existing one. |
| `registry__list_traces` | Read | List account runs visible to the caller. |
| `registry__get_trace` | Read | Read one account run. |
| `registry__get_trace_spans` | Read | Read the spans inside one account run. |

## Agent tools

| Wire name | Effect | Use |
| - | - | - |
| `agent__create_bulk` | Write | Create several Agents without returning their credentials. |
| `agent__list_traces` | Read | List runs for an Agent or coding harness. |
| `agent__get_trace` | Read | Read one Agent or harness run. |
| `agent__get_trace_spans` | Read | Read spans for one Agent or harness run. |
| `agent__query_trace_stats` | Read | Aggregate Agent or harness run statistics. |
| `agent__list_session_traces` | Read | Find traces linked to a recorded session. |

## Skill tools

| Wire name | Effect | Use |
| - | - | - |
| `skill__create` | Write | Create one or more Skills from uploaded ZIP bundles. |
| `skill__list_traces` | Read | List runs that invoked a Skill. |
| `skill__get_trace` | Read | Read one Skill run. |
| `skill__get_trace_spans` | Read | Read the spans inside one Skill run. |
| `skill__query_trace_stats` | Read | Aggregate statistics over a Skill's runs. |

## File tools

| Wire name | Effect | Use |
| - | - | - |
| `file__request_upload` | Write | Mint a short-lived upload URL and upload key. |

## Ontology tools

| Wire name | Effect | Use |
| - | - | - |
| `ontology__get_context` | Read | Retrieve a bounded relevant subgraph from one exact release. |
| `ontology__get_export` | Read | Export one exact release through a versioned adapter. |
| `ontology__get_import_preview` | Read | Parse an interchange document into a non-mutating proposal and loss report. |
| `ontology__get_query_plan` | Read | Resolve a bounded SPARQL SELECT query into a release-pinned logical plan. |

## Downstream tools

Registered downstream MCP servers contribute a dynamic, account-specific
catalog. Gateway advertises each tool as `<server-slug>__<tool-name>`.
Colliding server names gain a numeric suffix, so copy the name from
`tools/list` rather than constructing it.

`GET /tools/mcp/v1/tools` reads the persisted discovery projection. That projection
is useful for inventory and reporting, but it can lag the executable catalog.
Use authenticated protocol `tools/list` to decide what an agent can call.

## Skill metadata is not a tool declaration

A Skill's `allowed-tools` frontmatter records dependencies. It does not
register those tools with Gateway. A tool named in Skill metadata is callable
only when Registry, an installed extension, or a registered downstream server
advertises it.