Tools overview

The MCP server exposes a set of tools that your AI assistant (Claude, ChatGPT, Gemini, etc.) can call to read Agent configuration, query metrics, pull conversation data, manage test cases and test runs, and propose changes to resources (knowledge articles, custom instructions, test cases). Each tool page documents parameters, response shape, and example usage.

Tool index

Configuration

Tool	What it does
`get_ada_configuration`	Fetches playbooks (summary), actions, custom instructions, and company description.

Metrics and conversations

Tool	What it does
`get_ada_metric`	Returns AR rate, CSAT, containment, and engaged conversation volume with conversation-level metadata.
`get_available_filters`	Lists supported filter types and operators for conversation queries.
`get_conversations`	Returns many conversations at `IDS_ONLY` or `SUMMARY` detail.
`get_conversation`	Returns the full transcript for a conversation ID.

Knowledge and coaching

Tool	What it does
`search_knowledge`	Searches existing knowledge articles.
`search_coaching`	Searches existing coaching entries.

Entity discovery

Tool	What it does
`list_entities`	Lists channels, topics, variables, playbooks, or handoffs configured for the Agent.

Testing

Tool	What it does
`get_test_cases`	Retrieves test cases, optionally filtered by name or grouped by channel.
`get_test_runs`	Retrieves test run results, pass/fail status, and evaluation rationale.
`get_test_run_quota`	Returns daily test run quota, usage, and remaining runs.

Writes

Tool	What it does
`propose_change`	Creates, updates, or deletes knowledge articles, custom instructions, test cases, and test runs through a two-phase flow.

Typical call patterns

Performance analysis: get_available_filters → get_conversations → get_conversation (for selected IDs) → get_ada_metric for aggregates.
Configuration review: get_ada_configuration → search_knowledge / search_coaching to cross-reference existing content.
Improvement recommendations: get_ada_metric (to find under-performing conversations) → get_conversation (for transcripts) → get_ada_configuration (to identify gaps) → propose_change to create or update knowledge or custom instructions.
Test-driven development: get_conversations (production samples) → propose_change with test_case → propose_change with test_run → get_test_runs to review results.