Tool reference

The 67 tools of the management MCP, grouped by area.

You don't need to call them by name: the chatbot picks the tool from what you ask. This reference shows what's possible.

Read tools are on every plan. Write tools only appear with full access (see Permissions and plans).

Agents#

Tool	Type	What it does
`list_agents`	Read	Lists the agents with their published version and latest saved one
`get_agent`	Read	Shows an agent: draft, versions, published version, and quality gate
`create_agent`	Write	Creates an agent with a default draft
`update_agent`	Write	Changes the name or description
`archive_agent`	Write	Archives an agent: it stops serving customers
`restore_agent`	Write	Restores an archived agent
`get_agent_draft`	Read	Shows the draft
`update_agent_draft`	Write	Changes only the given settings of the draft
`validate_agent_draft`	Read	Checks the draft against the workspace and the plan
`save_agent_version`	Write	Saves the draft as a new version
`publish_agent_version`	Write	Publishes a version to your customers
`list_agent_versions`	Read	Lists the saved versions
`get_agent_version`	Read	Shows a version
`get_system_prompt`	Read	Shows the final prompt of the draft or a version
`get_context_usage`	Read	Estimates how much context each part takes
`list_models`	Read	Lists the models with their context window and price in credits
`get_model_support`	Read	Shows which efforts, caching, and credentials a model supports
`get_plan`	Read	Shows your plan's limits, features, and credits

Tool	Type	What it does
`list_instruction_sets`	Read	Lists the instruction sets
`get_instruction_set`	Read	Shows one with its latest version in full
`create_instruction_set`	Write	Creates new instructions (version 1)
`add_instruction_version`	Write	Adds a version with the new text
`validate_instruction_set`	Read	Checks for errors and warnings
`archive_instruction_set`	Write	Archives a set

Tool	Type	What it does
`list_eval_datasets`	Read	Lists the test datasets
`get_eval_dataset`	Read	Shows a dataset
`create_eval_dataset`	Write	Creates 1 to 200 test cases
`add_eval_dataset_version`	Write	Adds a version with the new list of cases
`archive_eval_dataset`	Write	Archives a set
`list_eval_suites`	Read	Lists the suites
`get_eval_suite`	Read	Shows a suite
`create_eval_suite`	Write	Creates a suite: cases, evaluators, and pass threshold
`add_eval_suite_version`	Write	Adds a version of the suite
`archive_eval_suite`	Write	Archives a suite
`run_evaluation`	Write	Evaluates a saved version with a suite (spends credits)
`get_evaluation`	Read	Shows the result, case by case
`list_evaluations`	Read	Lists an agent's evaluations
`cancel_evaluation`	Write	Cancels a running evaluation
`set_quality_gate`	Write	Requires a passed suite before publishing
`remove_quality_gate`	Write	Removes that requirement

Tool	Type	What it does
`list_http_tools`	Read	Lists the HTTP tools
`get_http_tool`	Read	Shows a tool
`create_http_tool`	Write	Creates a tool that calls an endpoint
`add_http_tool_version`	Write	Adds a version
`test_http_tool`	Read	Shows the request it would send, without sending it
`archive_http_tool`	Write	Archives a tool

Tool	Type	What it does
`list_mcp_connections`	Read	Lists the connections
`get_mcp_connection`	Read	Shows a connection and its last test
`create_mcp_connection`	Write	Connects a server without authentication. If it needs a token or OAuth, returns the console link
`test_mcp_connection`	Write	Tests the connection and discovers its tools
`list_mcp_connection_tools`	Read	Lists the server's tools
`archive_mcp_connection`	Write	Archives a connection

Tool	Type	What it does
`list_skills`	Read	Lists the workspace's skills
`get_skill`	Read	Shows a skill
`create_skill`	Write	Creates a skill
`add_skill_version`	Write	Adds a version
`archive_skill`	Write	Archives a skill
`search_skill_catalog`	Read	Searches the Productos AI catalog
`get_catalog_skill`	Read	Shows a catalog skill
`copy_catalog_skill`	Write	Copies a catalog skill into the workspace

Tool	Type	What it does
`list_knowledge_bases`	Read	Lists the bases and how many documents are indexed
`get_knowledge_base`	Read	Shows a base with its documents
`create_knowledge_base`	Write	Creates a base
`add_knowledge_text`	Write	Adds a Markdown document
`start_knowledge_ingestion`	Write	Indexes the documents
`get_knowledge_ingestion`	Read	Shows the indexing progress
`archive_knowledge_base`	Write	Archives a base