Tool Reference

The Bland MCP server exposes two kinds of tools, plus 3 prompts on API key connections:

Dedicated tools cover the common workflows with typed inputs. Some render inline UI in clients that support it, such as the call-log inspector, eval scorecards, and the analytics panel.
API Reference tools are one tool per page of the API Reference. Each takes the inputs on its linked page, sends the request as your organization, and returns the API’s JSON under data.

Every tool is scoped to one organization: the one that owns the API key, or the one you chose when you signed in with Bland. Connect with a full-access API key or by signing in. Restricted API keys can’t connect to the MCP server. Access is each tool’s annotation. Read tools never change anything. Write tools have side effects. Destructive tools delete things, spend money, or touch live traffic, so your client should ask you before it runs one. The server doesn’t ask for confirmation itself, so keep your client’s tool-approval prompts on. Sign-in is the scope a sign-in connection needs to call the tool. mcp:write also grants mcp:read. Not offered means the tool isn’t listed over sign-in; use an API key connection for it. None means the tool needs no scope. A full-access API key reaches every tool without scope checks.

Dedicated tools

Agents

Full lifecycle for the Bland Agents platform: versioned, branched, and promoted through dev → staging → production environments.

For Bland Agents builds, migrations, snapshot audits, and simulations, use the Norm plugin (norm@bland). The separate bland@bland plugin covers the v1 pathway workspace.

Tool	Access	Sign-in	Description
`create_agent`	Write	`mcp:write`	Create a new agent plus its dev, staging, and production environments, all unpinned.
`list_agents`	Read	`mcp:read`	List the organization’s agents, most recently updated first. Deleted agents are excluded.
`get_agent`	Read	`mcp:read`	Fetch one agent by ID, including which version staging and production are pinned to.
`rename_agent`	Write	`mcp:write`	Rename an agent.
`delete_agent`	Destructive	`mcp:write`	Soft-delete an agent. Its version history and deployment records stay.
`save_agent_version`	Write	`mcp:write`	Save the agent’s full `{behavior, settings, contact}` config as an immutable version snapshot.
`list_agent_versions`	Read	`mcp:read`	List an agent’s saved version history (metadata only), newest first.
`get_latest_agent_version`	Read	`mcp:read`	Fetch the latest saved version, including its full config snapshot.
`get_agent_version`	Read	`mcp:read`	Fetch a specific published version by semver, including its full snapshot.
`list_agent_environments`	Read	`mcp:read`	List the dev, staging, and production environments and the semver each is pinned to.
`publish_agent`	Destructive	`mcp:write`	Publish to staging: a new `snapshot`, an existing `versionId`, or the dev head. A new version gets the next semver.
`promote_agent`	Destructive	`mcp:write`	Promote to production: what staging points at, or a specific `versionId`. Enforces the saved evaluation policy unless `force` is set.
`rollback_agent`	Destructive	`mcp:write`	Point production back at a version that was deployed there before, keeping its original semver.
`list_agent_deployments`	Read	`mcp:read`	List an environment’s deployment history (production by default), with rollback candidates flagged.
`get_agent_check_config`	Read	`mcp:read`	Fetch the judges and scenarios that gate promotions into an environment.
`set_agent_check_config`	Destructive	`mcp:write`	Create or replace the production check config: which evals score a candidate, which are required, and which scenarios run.
`delete_agent_check_config`	Destructive	`mcp:write`	Delete an environment’s check config.
`start_agent_check_run`	Destructive	`evals:run`	Score simulated conversations against a candidate version with the environment’s judges. Bills the organization. Never moves environment pointers.
`list_agent_check_runs`	Read	`mcp:read`	List an environment’s check runs, newest first, with per-eval verdicts.
`get_agent_check_run`	Read	`mcp:read`	Fetch one check run, including its per-eval verdicts and simulation set.
`cancel_agent_check_run`	Destructive	`mcp:write`	Cancel a pending or running check run.
`list_agent_variables`	Read	`mcp:read`	List an environment’s variables for an agent.
`set_agent_variable`	Destructive	`mcp:write`	Create or update one environment variable. Set `isSecret` to store the value as a secret.
`delete_agent_variable`	Destructive	`mcp:write`	Delete one environment variable.
`list_agent_branches`	Read	`mcp:read`	List an agent’s open branches (isolated dev workspaces).
`create_agent_branch`	Write	`mcp:write`	Branch off the agent’s current dev head.
`delete_agent_branch`	Destructive	`mcp:write`	Abandon an open branch. Its saved versions stay.
`rebase_preview_agent_branch`	Read	`mcp:read`	Dry-run a branch merge: a three-way diff of what merges cleanly and what conflicts.
`rebase_agent_branch`	Destructive	`mcp:write`	Rebase a branch onto the dev head with per-conflict resolutions.
`merge_agent_branch`	Destructive	`mcp:write`	Merge an open branch into dev (fast-forward only) and archive the branch.

More agent tools (inbound numbers, experiments, dispositions, migration) are under API Reference tools.

Calls

Tool	Access	Sign-in	Description
`create_call`	Destructive	`calls:create`	Place an outbound call with the common fields only: `phoneNumber`, `task`, `pathwayId`, `voice`, `from`, and `firstSentence`. Omit `voice` to use Karen.
`get_call_log`	Read	`mcp:read`	Fetch a single call: its transcript, recording, summary, and metadata.
`wait_for_call`	Read	`mcp:read`	Wait up to 45 seconds (30 by default) for a call to finish, then return its call log. Use it instead of polling `get_call_log`.
`stop_call`	Destructive	`calls:stop`	End a call in progress, or cancel one that is still queued.
`list_voices`	Read	`mcp:read`	List Bland’s curated voices, the recommended set for new calls, personas, numbers, and pathways. Any voice the user names still works.

Choose the call resource before invoking a tool

Follow Choose what runs your call: resolve whether the user wants a dashboard agent, persona, or pathway before sending. Honor an established choice, verify the exact ID and environment or version, and ask before changing routes. create_call does not accept agent_id, agent_version, persona_id, or pathway-version selectors. Do not replace an agent choice with pathwayId just to fit this schema. Use send_call instead. It takes every Send Call field in the same snake_case shape, needs the same calls:create scope over sign-in, and runs the call through exactly one of agent_id, persona_id, pathway_id, or task. For example, after verifying the destination and agent ID:

{
  "phone_number": "+14155550100",
  "agent_id": "00000000-0000-4000-8000-000000000001",
  "agent_version": "production"
}

Over an API key connection, call_bland_api with method: "POST" and path: "/v1/calls" takes the same body. Use list_agents, get_agent, and list_agent_environments to verify an agent and its deployment before calling. If the required tool or route is unavailable, explain the limitation and ask the user how to proceed. Do not silently use a persona or pathway instead. After submission, retain the call_id, check the actual source and destination numbers and resource/version association, and verify agent calls under the intended agent’s Conversations view. The create_call result and call-log widget alone do not establish agent attribution. Report a successful call and an attribution problem separately, without placing a duplicate call.

Analytics

Tool	Access	Sign-in	Description
`query_analytics`	Read	`mcp:read`	Structured call-analytics queries (metrics, grouping, filters, and time ranges) returning aggregated results.

Evals

Tool	Access	Sign-in	Description
`list_eval_agents`	Read	`mcp:read`	List the organization’s eval agents (LLM judges) with their latest-run summaries.
`get_eval_agent`	Read	`mcp:read`	Fetch one eval agent: criterion prompt, scoring levels, targets, weight, and modality.
`get_eval_run`	Read	`mcp:read`	Fetch one eval run and its per-atom verdicts.
`create_eval_run`	Destructive	`evals:run`	Enqueue scoring of a batch of calls against eval agents or a saved workbench setup.

Knowledge and code

Tool	Access	Sign-in	Description
`upload_knowledge_file`	Write	`mcp:write`	Add a file to the knowledge base from an `https` URL (20 MB max; pdf, docx, xlsx, txt, md, csv, or json). Indexing runs in the background.
`save_code_snippet`	Destructive	`mcp:write`	Create a JavaScript custom-code snippet, or save new code as an existing snippet’s next version. Tools that don’t pin `snippet_version` run the latest version.

Billing

Both tools take a shared_payment_token (spt_...) that the owner approves in their agent wallet. They need an organization API key and aren’t offered over sign-in. If purchases aren’t enabled for your organization, they return FEATURE_DISABLED.

Tool	Access	Sign-in	Description
`buy_credits`	Destructive	Not offered	Buy prepaid credits, $5 to $100 per purchase. The caller must be an owner, admin, or operator. A settled purchase unlocks international calling for an organization without the Agent Phone Plan.
`buy_phone_plan`	Destructive	Not offered	Add the Agent Phone Plan to the organization, with an optional US `area_code` for its number. The caller must be an owner or admin.

Documentation

search_bland, query_docs_filesystem_bland, and submit_feedback are passed through from the docs site’s own MCP server (https://docs.bland.ai/mcp). If that server is unreachable, they’re missing from the list.

Tool	Access	Sign-in	Description
`search_bland_docs`	Read	`mcp:read`	Search docs.bland.ai and return matching pages with title, URL, and description.
`get_bland_doc`	Read	`mcp:read`	Fetch the full markdown of one docs page (for example `api-v1/post/calls`).
`search_bland`	Read	None	Search the Bland knowledge base for guides, code examples, and API references.
`query_docs_filesystem_bland`	Read	None	Read-only shell-like queries over a virtual filesystem of the docs pages.
`submit_feedback`	Write	None	Report a documentation problem to the docs team.

API passthrough

Over an API key connection, the passthrough pair reaches the entire Bland REST API, including fields a dedicated tool leaves out. Sign-in connections don’t get it; use the dedicated and API Reference tools instead.

Tool	Access	Sign-in	Description
`bland_api_get`	Read	Not offered	GET any Bland API endpoint (for example `/v1/calls` with query params).
`call_bland_api`	Destructive	Not offered	Call any mutating Bland endpoint (POST, PUT, PATCH, or DELETE).

Setup

Tool	Access	Sign-in	Description
`get_bland_mcp_setup`	Read	None	Return connection guidance for the Bland MCP server.

Pathways (Legacy)

These tools work with legacy Conversational Pathways. For new builds, use the Agents tools above.

Tool	Access	Sign-in	Description
`validate_pathway`	Read	`mcp:read`	The authoritative Bland pathway compiler. Pass the raw graph you’re about to commit and get structural and runtime-contract errors before saving. Pass `baseline` to see only what your edit introduced.
`get_pathway_schema`	Read	`mcp:read`	The authoritative allowed shape and enums for a structured pathway surface, node, or edge.
`get_pathway_context`	Read	`mcp:read`	Deep per-node and per-edge execution semantics, the same summaries Bland’s server-side pathway agent uses.

API Reference tools

Each tool below matches one API Reference page, and most names link to it. Its inputs are that page’s path, query, and body parameters. Endpoints that already have a dedicated tool, such as Stop Call (stop_call) and Call Details (get_call_log), aren’t repeated here. Sign-in connections reach an allowlist of everyday resources. A tool is Not offered over sign-in when its endpoint is outside that allowlist, or when it moves money, mints a key, session, or token, controls live calls, sends stored credentials to a third party, or runs model work Bland pays for.

Agent inbound numbers

Tool	Access	Sign-in	Description
`get_agent_number_binding`	Read	`mcp:read`	Check which phone numbers are attached to an agent and whether each one answers on the production or staging environment.
`attach_agent_numbers`	Write	`mcp:write`	Route inbound calls on your phone numbers to an agent.
`detach_agent_numbers`	Write	`mcp:write`	Stop routing inbound calls on phone numbers to an agent.

Agent versions

Tool	Access	Sign-in	Description
`get_agent_scenario_history`	Read	`mcp:read`	List how one scenario changed across saved versions.

Agent checks

Tool	Access	Sign-in	Description
`remove_agent_check_eval`	Destructive	`mcp:write`	Drop one eval from an environment’s promotion checks.

Agent experiments

Tool	Access	Sign-in	Description
`create_agent_experiment`	Write	`mcp:write`	Start an A/B experiment on a variant version.
`list_agent_experiments`	Read	`mcp:read`	List an agent’s experiments, newest first.
`get_agent_experiment`	Read	`mcp:read`	Retrieve one experiment and its live progress.
`update_agent_experiment`	Write	`mcp:write`	Adjust an active experiment’s split or conditions.
`stop_agent_experiment`	Write	`mcp:write`	End an active experiment and return traffic to baseline.

Agent identity

Tool	Access	Sign-in	Description
`get_agent_identity`	Read	Not offered	Read the agent’s branded contact card and its delivery settings.
`update_agent_identity`	Write	Not offered	Change the agent’s contact card fields and delivery settings.
`get_agent_contact_card`	Read	Not offered	Download the agent’s contact card as a vCard file.
`send_agent_test_contact_card`	Destructive	Not offered	Send the agent’s contact card to one number to check how it arrives.

Agent dispositions

Tool	Access	Sign-in	Description
`list_agent_dispositions`	Read	`mcp:read`	List the dispositions defined on an agent.
`delete_agent_disposition`	Destructive	`mcp:write`	Retire a disposition so it stops classifying new calls.
`create_agent_disposition`	Write	`mcp:write`	Create a disposition with an initial draft definition.
`get_agent_disposition`	Read	`mcp:read`	Retrieve a disposition’s summary and lifecycle state.
`get_agent_disposition_draft`	Read	`mcp:read`	Retrieve a disposition’s live draft definition.
`update_agent_disposition_draft`	Write	`mcp:write`	Replace a disposition’s draft definition.
`list_agent_disposition_versions`	Read	`mcp:read`	List a disposition’s published versions, newest first.
`get_agent_disposition_version`	Read	`mcp:read`	Retrieve one published disposition version.
`estimate_agent_disposition`	Write	`mcp:write`	Estimate a disposition’s per-call cost before publishing.
`publish_agent_disposition`	Destructive	`mcp:write`	Publish a disposition draft as an immutable version.
`disable_agent_disposition`	Destructive	`mcp:write`	Stop a disposition from running on new calls.
`list_agent_call_disposition_runs`	Read	`mcp:read`	Get the latest production disposition results for up to 100 of an agent’s calls, with each run’s values, status, and delivery timing.
`list_disposition_test_runs`	Read	`mcp:read`	List test runs for a disposition.
`create_disposition_test_run`	Destructive	`evals:run`	Run a disposition against completed production calls.
`get_disposition_test_run`	Read	`mcp:read`	Retrieve a test run with live progress counts.
`list_disposition_test_run_rows`	Read	`mcp:read`	Page through a test run’s per-call results.
`export_disposition_test_run_csv`	Read	`mcp:read`	Download a test run’s results as CSV.
`align_disposition_test_run`	Destructive	Not offered	Propose definition edits from a test run’s corrections.
`set_disposition_test_annotation`	Write	`mcp:write`	Record a corrected value for one call in a test run.
`get_disposition_test_annotation`	Read	`mcp:read`	List correction revisions for one call and value.
`list_disposition_extractors`	Read	`mcp:read`	List the extractors available to an agent’s dispositions.
`create_disposition_extractor`	Write	`mcp:write`	Create an extractor with its first editable version.
`get_disposition_extractor`	Read	`mcp:read`	Retrieve an extractor with its draft and published versions.
`get_disposition_extractor_version`	Read	`mcp:read`	Retrieve one version of an extractor.
`update_disposition_extractor_draft`	Write	`mcp:write`	Edit an extractor’s editable draft version.
`publish_disposition_extractor`	Write	`mcp:write`	Publish an extractor draft as an immutable version.
`get_disposition_judge_catalog`	Read	`mcp:read`	List the judges a disposition can pin for this agent.
`get_disposition_judge_version`	Read	`mcp:read`	Retrieve one version of a judge’s rubric.
`get_disposition_variable_catalog`	Read	`mcp:read`	List the agent variables a disposition can re-extract.
`get_disposition_model_profile_catalog`	Read	`mcp:read`	List the model profiles available to extractors.
`get_disposition_extractor_catalog`	Read	`mcp:read`	List the published extractors a disposition can pin.

Agent library

Tool	Access	Sign-in	Description
`list_library_scenarios`	Read	`mcp:read`	List reusable scenarios from your other agents.
`list_archived_nodes`	Read	`mcp:read`	List nodes archived from your organization’s agents.
`delete_archived_node`	Destructive	`mcp:write`	Permanently remove an entry from the node archive.
`archive_agent_node`	Write	`mcp:write`	Copy a node from an agent into the organization archive.

Migration to agents

Tool	Access	Sign-in	Description
`migrate_pathway_to_agent`	Destructive	Not offered	Plan and build an agent from a legacy pathway graph.
`migrate_persona_to_agent`	Write	Not offered	Build an agent from a persona and its attached pathways.
`migrate_pathways_to_agents`	Write	Not offered	Convert stored or inline pathways into agent scenarios.
`start_pathway_translation`	Write	Not offered	Start an async job translating pathways into scenarios.
`get_pathway_translation_run`	Read	Not offered	Poll the status and result of a translation job.
`cancel_pathway_translation_run`	Write	Not offered	Cancel a pending or running translation job.

Calls API

Tool	Access	Sign-in	Description
`send_call`	Destructive	`calls:create`	Send an AI phone call with a custom objective and actions.
`analyze_call`	Destructive	Not offered	Analyze a call with AI, using questions and goals you supply.
`get_call_listen_url`	Write	Not offered	Initiate a live listen session for an active call.
`get_call_transcript_stream_url`	Write	Not offered	Get a short-lived, ready-to-connect WebSocket URL for a call’s live post-transfer transcript stream.
`stop_all_active_calls`	Destructive	Not offered	End all active phone calls on your account.
`transfer_active_call`	Destructive	Not offered	Transfer an in-progress call to a different phone number.
`list_calls`	Read	`mcp:read`	Returns a set of metadata for each call dispatched by your account.
`list_active_calls`	Read	`mcp:read`	Retrieve all currently active calls for your organization.
`list_call_events`	Read	Not offered	Retrieve stream of events that occurred during the call.
`get_corrected_transcript`	Read	Not offered	Get the corrected transcript from your call’s recording.
`get_post_call_webhook`	Read	`mcp:read`	Get the post call webhook data for a specific call.
`create_post_call_webhooks`	Destructive	`mcp:write`	Create and send post call webhooks for specified calls.
`resend_post_call_webhooks`	Destructive	`mcp:write`	Resend post call webhooks for specified calls.

Batches

Tool	Access	Sign-in	Description
`create_batch`	Destructive	`calls:create`	Create a new batch of calls using direct input.
`list_batches`	Read	`mcp:read`	Retrieve a list of all batches created by your organization.
`get_batch`	Read	`mcp:read`	Retrieve metadata for a specific batch.
`get_batch_logs`	Read	`mcp:read`	Retrieve logs for a specific batch.
`stop_batch`	Destructive	`calls:stop`	Stop a batch that is currently running or scheduled.

SMS

Tool	Access	Sign-in	Description
`list_sms_numbers`	Read	`mcp:read`	Retrieve all phone numbers configured for SMS, along with their configuration and conversation counts.
`update_sms_number_config`	Write	`mcp:write`	Update the SMS configuration for a phone number owned by the authenticated user.
`send_sms`	Destructive	`mcp:write`	Send an SMS message from an agent to a user.
`send_sms_batch`	Destructive	`mcp:write`	Send SMS messages to a large list of recipients from a pre-uploaded CSV file.
`create_sms_conversation`	Write	`mcp:write`	Create an SMS conversation with specific pathway state without triggering immediate message sending.
`list_sms_conversations`	Read	`mcp:read`	Retrieve a paginated list of SMS conversations for the authenticated user, with filtering, sorting, and metadata.
`get_sms_conversation`	Read	`mcp:read`	Retrieve a single SMS conversation and its messages by its ID.
`update_sms_conversation`	Write	`mcp:write`	Updates properties of an existing SMS conversation.
`hand_off_sms_conversation`	Write	`mcp:write`	Pause the agent on a conversation so a human can take over the thread.
`send_sms_conversation_message`	Destructive	`mcp:write`	Send a message on an existing conversation from the agent number, as a human rep or as the agent.
`resume_sms_conversation`	Write	`mcp:write`	Hand a conversation back to the agent after a human handoff.
`end_sms_conversation`	Write	`mcp:write`	Close a conversation for good, typically when a human rep is done with it.
`delete_sms_conversation_messages`	Destructive	`mcp:write`	Delete messages from an SMS conversation.
`delete_sms_conversation`	Destructive	`mcp:write`	Delete a specific conversation and all of its associated messages.
`get_sms_conversation_webhook`	Read	`mcp:read`	Retrieve the webhook delivery log for an SMS conversation, including the payload, URL, response codes, and timing.
`analyze_sms_conversation`	Write	Not offered	Answer questions and extract information from an SMS conversation.

Numbers

Tool	Access	Sign-in	Description
`purchase_phone_number`	Destructive	Not offered	Purchase a new phone number (inbound/outbound).
`update_inbound_number`	Write	`mcp:write`	Update your inbound agent’s settings, prompt and other details.
`update_inbound_number_label`	Write	Not offered	Update the label shown for one of your inbound phone numbers.
`list_phone_numbers`	Read	`mcp:read`	Retrieves a list of all inbound phone numbers configured for your account, along with their associated settings.
`get_phone_number`	Read	`mcp:read`	Retrieve settings for your inbound phone number.
`create_inbound_session`	Write	Not offered	Create a session token that can be used to pass variables to inbound calls via SIP headers.
`get_number_voice_pool`	Read	`mcp:read`	List the voices in an inbound number’s voice pool.
`set_number_voice_pool`	Write	`mcp:write`	Replace the voice pool on an inbound number.

Blocked numbers

Tool	Access	Sign-in	Description
`create_block_rules`	Write	Not offered	Create one or more block rules to prevent specific phone numbers from reaching your inbound numbers.
`list_block_rules`	Read	`mcp:read`	Retrieve block rules associated with your inbound numbers.
`get_block_rule`	Read	`mcp:read`	Retrieve the details of a specific block rule.
`delete_block_rule`	Destructive	Not offered	Permanently delete a block rule.
`edit_block_rule`	Write	Not offered	Update attributes of an existing block rule.

Custom dialing pools

Tool	Access	Sign-in	Description
`get_custom_dialing_pool`	Read	Not offered	Retrieve a specific custom dialing pool by its ID.
`list_custom_dialing_pools`	Read	Not offered	Retrieve all custom dialing pools for your organization.
`create_custom_dialing_pool`	Write	Not offered	Create a new custom dialing pool with phone numbers for optimized call routing.
`update_custom_dialing_pool`	Write	Not offered	Update an existing custom dialing pool with new phone numbers or credentials.

SIP trunks

Tool	Access	Sign-in	Description
`get_sip_config`	Read	Not offered	Retrieve the current SIP configuration for a phone number.
`list_sip_numbers`	Read	Not offered	List all phone numbers with SIP configurations for your organization.
`attach_sip_config`	Write	Not offered	Attach SIP configuration to one or more phone numbers for inbound and/or outbound routing.
`update_sip_direction`	Write	Not offered	Update an existing SIP routing direction for a phone number.
`detach_sip_config`	Destructive	Not offered	Remove SIP configuration from a phone number for specified directions.
`parse_sip_destination`	Write	Not offered	Parse and validate a raw SIP destination string into its host, port, user, and transport before attaching it to a trunk.
`discover_sip_endpoint`	Write	Not offered	Auto-discover optimal SIP connection settings by probing your endpoint with DNS resolution and SIP OPTIONS.
`get_sip_discovery_status`	Read	Not offered	Poll the status and results of a SIP endpoint discovery session.
`send_sip_test_call`	Destructive	Not offered	Place a live test call to your SIP endpoint to verify connectivity.
`get_sip_test_call_status`	Read	Not offered	Poll the status and trace of a SIP test call.
`get_sip_firewall_ips`	Read	Not offered	Get Bland’s static SIP signaling and media IP addresses and ports for your region.
`get_sip_outbound_setup`	Read	Not offered	Get Bland’s SIP server details and PBX configuration instructions for outbound call routing.
`generate_sip_password`	Read	Not offered	Generate a strong, PBX-safe SIP password for register-based authentication, at a configurable length.
`get_sip_trunk_health`	Read	Not offered	Get the health status of a SIP trunk by probing the endpoint with SIP OPTIONS.
`list_sip_calls`	Read	Not offered	Retrieve SIP call logs for a phone number with cursor-based pagination.
`update_sip_advanced_config`	Write	Not offered	Update advanced SIP configuration including failover, codecs, and alerts.
`check_number_portability`	Read	Not offered	Check whether phone numbers are eligible for porting to Bland.
`initiate_port_request`	Destructive	Not offered	Submit a number porting request to transfer phone numbers from another carrier to Bland.
`list_port_requests`	Read	Not offered	List all number porting requests for your organization.
`cancel_port_request`	Destructive	Not offered	Cancel an in-progress number porting request.

Custom Twilio accounts

Tool	Access	Sign-in	Description
`create_encrypted_key`	Destructive	Not offered	Integrate your own Twilio account with Bland.
`delete_encrypted_key`	Destructive	Not offered	Disable an encrypted key for a Twilio account integration.
`upload_inbound_numbers`	Destructive	Not offered	Add inbound numbers to Bland from your own Twilio account.
`delete_inbound_number`	Destructive	Not offered	Remove an inbound number that was uploaded through your own Twilio account.

Knowledge bases

Tool	Access	Sign-in	Description
`create_knowledge_base_from_text`	Write	`mcp:write`	Creates a new knowledge base item from direct text input.
`create_knowledge_base_from_websites`	Write	`mcp:write`	Creates a new knowledge base item by scraping content from web URLs.
`list_knowledge_bases`	Read	`mcp:read`	Retrieves a paginated list of knowledge base items for the authenticated organization.
`get_knowledge_base`	Read	`mcp:read`	Retrieves a specific knowledge base item by ID.
`update_knowledge_base`	Write	`mcp:write`	Updates a knowledge base item’s name and/or description.
`delete_knowledge_base`	Destructive	`mcp:write`	Soft deletes a knowledge base item by setting its status to ‘DELETED’.
`discover_sitemap_urls`	Write	`mcp:write`	Discovers URLs from a website’s sitemap for web scraping.
`chat_with_knowledge_base`	Write	Not offered	Performs a conversational query against knowledge base items with context.

Custom tools

Tool	Access	Sign-in	Description
`create_tool`	Write	Not offered	Create a new tool.
`update_tool`	Write	Not offered	Update an existing tool.
`list_tools`	Read	Not offered	Retrieve a paginated list of your tools.
`list_tool_execution_logs`	Read	Not offered	Retrieve per-execution logs for your tools with optional filtering and pagination.
`get_tool_execution_stats`	Read	Not offered	Retrieve aggregated execution statistics for your tools, grouped and filtered by integration, status, date, and more.

Contacts

Tool	Access	Sign-in	Description
`list_contacts`	Read	`mcp:read`	List all contacts for your organization.
`find_contact`	Write	`mcp:write`	Find a contact by phone number, email, or external ID.
`get_contact`	Read	`mcp:read`	Retrieve a contact by its unique ID, including all identifiers, contact_memories (per persona/agent), and memory entities.
`resolve_contact`	Write	`mcp:write`	Find an existing contact or create a new one.
`update_contact`	Write	`mcp:write`	Update an existing contact’s information.
`merge_contacts`	Destructive	`mcp:write`	Merge two contacts into one.

Memory

Tool	Access	Sign-in	Description
`create_contact_memory`	Write	`mcp:write`	Create a memory record for a contact scoped to a persona or agent number.
`get_contact_memory`	Read	`mcp:read`	Retrieve a contact’s full memory record by ID, including their conversation summary, structured facts, recent messages, and open items.
`list_contact_memory_messages`	Read	`mcp:read`	Retrieve the recent message history for a contact memory.
`update_contact_memory_summary`	Write	`mcp:write`	Overwrite the conversation summary for a contact memory.
`update_contact_memory_facts`	Write	`mcp:write`	Merge new facts into a contact’s memory record.
`get_memory_context`	Read	Not offered	Retrieve the full memory context for a contact scoped to a persona or agent number.
`get_memory_changes`	Read	`mcp:read`	Retrieve what memory was captured or updated during a specific call or SMS conversation.
`reset_contact_memory`	Destructive	`mcp:write`	Permanently delete all memory for a contact scoped to a specific persona or agent number.
`set_memory_enabled`	Write	`mcp:write`	Enable or disable memory for a pathway or persona version.
`get_agent_memory_schema`	Read	`mcp:read`	Read what the agent remembers about each contact between calls.
`infer_agent_memory_schema`	Write	Not offered	Generate a suggested memory schema by reading the agent.

Agent testing

Tool	Access	Sign-in	Description
`create_test_scenario`	Write	`mcp:write`	Create a new test scenario with assertions for a pathway or persona.
`list_test_scenarios`	Read	`mcp:read`	List all test scenarios for your organization.
`get_test_scenario`	Read	`mcp:read`	Retrieve a specific test scenario with its assertions.
`update_test_scenario`	Write	`mcp:write`	Update an existing test scenario.
`delete_test_scenario`	Destructive	`mcp:write`	Delete a test scenario and all related data.
`list_test_templates`	Read	Not offered	List out-of-box test scenario templates.
`clone_test_template`	Write	`mcp:write`	Clone a template to create a new scenario bound to your pathway or persona.
`generate_test_scenario_from_call`	Write	Not offered	Generate a test scenario from a historical call transcript using AI analysis.
`run_test_scenario`	Destructive	`evals:run`	Execute a single test scenario.
`batch_run_test_scenarios`	Destructive	`evals:run`	Execute multiple test scenarios as a batch.
`list_test_runs`	Read	`mcp:read`	List test runs with optional filtering and pagination.
`get_test_run`	Read	`mcp:read`	Retrieve detailed results for a specific test run including chat history, assertion results, and scores.
`get_test_batch`	Read	`mcp:read`	Retrieve a test batch and all its runs.
`analyze_failed_test_run`	Write	Not offered	Run AI-powered analysis on a failed test run to get root cause and fix suggestions.
`get_test_analytics`	Read	`mcp:read`	Get basic testing analytics for a pathway over a time window.
`get_enhanced_test_analytics`	Read	`mcp:read`	Get enhanced testing analytics including node failure heatmap, weakest link, trend analysis, and Sankey flow data.
`create_simulation_set`	Destructive	`evals:run`	Create a simulation set that runs each scenario multiple times to detect flaky behavior.
`get_simulation_set`	Read	`mcp:read`	Retrieve a simulation set with its statistics including pass rates, flakiness detection, and score distributions.
`start_tornado_session`	Destructive	`evals:run`	Start an iterative fix loop that runs tests, analyzes failures, applies fixes, and retests until all scenarios pass.
`get_active_tornado_session`	Read	`mcp:read`	Get the currently active tornado session for a pathway.
`get_tornado_status`	Read	`mcp:read`	Get detailed progress for a tornado session including iteration data and fix plans.
`cancel_tornado_session`	Write	`mcp:write`	Cancel a running tornado session.

Evals API

Tool	Access	Sign-in	Description
`get_evals_status`	Read	`mcp:read`	Check whether the Evals feature is enabled for your organization.
`create_eval_agent`	Write	`mcp:write`	Create a new eval agent, optionally seeded from a template.
`update_eval_agent`	Write	`mcp:write`	Update an eval agent’s metadata or repoint its active version.
`delete_eval_agent`	Destructive	`mcp:write`	Soft-delete an eval agent.
`publish_eval_agent`	Write	`mcp:write`	Publish the agent’s current draft as a new active version.
`list_eval_agent_versions`	Read	`mcp:read`	List all versions of an eval agent.
`create_eval_agent_version`	Write	`mcp:write`	Fork a new editable draft version of an eval agent.
`get_eval_agent_version`	Read	`mcp:read`	Fetch one version of an eval agent.
`update_eval_agent_version`	Write	`mcp:write`	Edit a draft version’s prompt, levels, targets, or weight.
`list_eval_runs`	Read	`mcp:read`	List eval runs in your organization, with optional filters.
`estimate_eval_run`	Write	`mcp:write`	Preview the size and cost of a run before starting it.
`cancel_eval_run`	Write	`mcp:write`	Cancel a pending, queued, or running eval run.
`list_eval_run_call_results`	Read	`mcp:read`	List the per-call results for an eval run.
`list_eval_run_agent_results`	Read	`mcp:read`	List every individual judge verdict in an eval run.
`list_eval_run_agent_snapshots`	Read	`mcp:read`	List the frozen agent configurations a run scored against.
`list_workbench_setups`	Read	`mcp:read`	List the workbench setups in your organization.
`create_workbench_setup`	Write	`mcp:write`	Create a new workbench setup with an empty draft version.
`get_workbench_setup`	Read	`mcp:read`	Fetch one workbench setup and its current draft version.
`update_workbench_setup`	Write	`mcp:write`	Update a workbench setup’s metadata or repoint its active version.
`delete_workbench_setup`	Destructive	`mcp:write`	Soft-delete a workbench setup.
`publish_workbench_setup`	Write	`mcp:write`	Publish the setup’s current draft as a new active version.
`list_workbench_setup_versions`	Read	`mcp:read`	List all versions of a workbench setup.
`create_workbench_setup_version`	Write	`mcp:write`	Fork a new editable draft version of a workbench setup.
`get_workbench_setup_version`	Read	`mcp:read`	Fetch one version of a workbench setup.
`update_workbench_setup_version`	Write	`mcp:write`	Edit a draft version’s attached agents, threshold, run mode, or default calls.
`list_eval_agent_templates`	Read	`mcp:read`	List the read-only library of shipped eval agent templates.
`get_eval_agent_template`	Read	`mcp:read`	Fetch a single shipped eval agent template by its key.
`list_eval_user_templates`	Read	`mcp:read`	List your organization’s saved eval agent templates.
`create_eval_user_template`	Write	`mcp:write`	Save a new eval agent template, from scratch or by snapshotting an existing eval agent.
`get_eval_user_template`	Read	`mcp:read`	Fetch one of your organization’s saved eval agent templates.
`update_eval_user_template`	Write	`mcp:write`	Update a saved eval agent template.
`delete_eval_user_template`	Destructive	`mcp:write`	Soft-delete a saved eval agent template.

Guard rails

Tool	Access	Sign-in	Description
`list_guard_rails`	Read	`mcp:read`	Retrieve a list of all guard rails in your organization.
`get_guard_rail`	Read	`mcp:read`	Retrieve a specific guard rail by ID.
`create_guard_rail`	Write	`mcp:write`	Create a new guard rail for compliance monitoring.
`update_guard_rail`	Write	`mcp:write`	Update an existing guard rail configuration.
`delete_guard_rail`	Destructive	`mcp:write`	Delete a guard rail.

Prompts library

Tool	Access	Sign-in	Description
`list_prompts`	Read	`mcp:read`	Retrieves all your saved prompts.
`get_prompt`	Read	`mcp:read`	Retrieves data for a specific prompt_id.
`create_prompt`	Write	`mcp:write`	Create and store a prompt for future use.

Widgets

Tool	Access	Sign-in	Description
`create_widget`	Write	`mcp:write`	Creates a new widget.
`list_widgets`	Read	`mcp:read`	Retrieves all widgets associated with your account.
`get_widget`	Read	`mcp:read`	Retrieves a specific widget by ID.
`update_widget`	Write	`mcp:write`	Updates an existing widget.
`list_widget_threads`	Read	`mcp:read`	Retrieves all conversation threads for a specific widget.
`send_widget_live_agent_message`	Destructive	`mcp:write`	Send a message from a live agent to the user, or end the conversation.
`list_widget_custom_components`	Read	`mcp:read`	List your widget custom components.
`create_widget_custom_component`	Write	`mcp:write`	Create a new custom component.
`update_widget_custom_component`	Write	`mcp:write`	Update a widget custom component.
`delete_widget_custom_component`	Destructive	`mcp:write`	Deletes a specific custom component by ID.

Live translation

Tool	Access	Sign-in	Description
`create_translation_session`	Write	Not offered	Create a real-time translation session and receive a WebSocket URL for streaming audio.
`get_translation_session`	Read	Not offered	Retrieve the status, duration, and billing details of a translation session.
`end_translation_session`	Destructive	Not offered	Terminate a translation session from the server side.

Triage

Tool	Access	Sign-in	Description
`create_triage_issue`	Write	`mcp:write`	Create a triage issue.
`list_triage_issues`	Read	`mcp:read`	List triage issues with filters and keyset pagination.
`get_triage_issue`	Read	`mcp:read`	Get a triage issue by ID.
`update_triage_issue`	Write	`mcp:write`	Update a triage issue.
`delete_triage_issue`	Destructive	`mcp:write`	Permanently delete a triage issue.
`attach_triage_resource`	Write	`mcp:write`	Attach a call, SMS conversation, or file to an issue.
`list_triage_resources`	Read	`mcp:read`	List resources attached to an issue.
`remove_triage_resource`	Destructive	`mcp:write`	Detach a resource from an issue.
`attach_triage_call`	Write	`mcp:write`	Attach a call to an issue.
`detach_triage_call`	Destructive	`mcp:write`	Detach a call from an issue.
`get_triage_affected_context`	Read	`mcp:read`	Calls, pathways, and personas affected by an issue.
`add_triage_flag`	Write	`mcp:write`	Flag a moment on an attached call.
`list_triage_flags`	Read	`mcp:read`	List flags on an issue.
`remove_triage_flag`	Destructive	`mcp:write`	Remove a flag from an issue.
`list_triage_flag_types`	Read	`mcp:read`	List reusable flag types in your org.
`list_triage_activity`	Read	`mcp:read`	Activity feed for an issue.
`add_triage_comment`	Write	Not offered	Comment on an issue.
`mark_triage_issue_viewed`	Write	`mcp:write`	Mark an issue as viewed by the calling user.
`list_triage_categories`	Read	`mcp:read`	List issue categories.
`create_triage_category`	Write	`mcp:write`	Create a custom issue category.
`link_triage_issue`	Write	`mcp:write`	Link an issue to another issue.
`list_triage_relations`	Read	`mcp:read`	List relations on an issue.
`remove_triage_relation`	Destructive	`mcp:write`	Remove a relation.
`list_triage_norm_agents`	Read	`mcp:read`	List triage agent profiles.
`create_triage_norm_session`	Write	Not offered	Create a Norm agent session on an issue.
`prompt_triage_norm_session`	Write	Not offered	Send a prompt to Norm.

Account

Tool	Access	Sign-in	Description
`get_account_details`	Read	Not offered	Get your account details.

API keys

Tool	Access	Sign-in	Description
`create_api_key`	Destructive	Not offered	Create a full-access or restricted API key for your organization.
`list_api_keys`	Write	Not offered	List your organization’s API keys and the permissions each one holds.
`rename_api_key`	Write	Not offered	Change the name of an API key.
`delete_api_key`	Destructive	Not offered	Revoke an API key.

Organizations

Tool	Access	Sign-in	Description
`create_organization`	Destructive	Not offered	Create a new organization.
`delete_organization`	Destructive	Not offered	Delete an organization.
`get_organization`	Read	Not offered	Retrieve details of an organization.
`list_organization_members`	Read	Not offered	Retrieve the list of members in an organization.
`get_org_current_service_version`	Read	Not offered	Retrieve the current version of a specified service for an organization.
`list_org_service_versions`	Read	Not offered	Retrieve a list of available versions for a specified service within an organization.
`update_org_service_version`	Destructive	Not offered	Update the current version of a specified service for an organization.
`update_organization_members`	Destructive	Not offered	Add or remove members from an organization.
`update_organization_member_permissions`	Destructive	Not offered	Modify the permissions of an existing member within an organization.
`update_organization_properties`	Write	Not offered	Modify specific properties of an organization.
`list_my_org_memberships`	Read	Not offered	Retrieve a list of organizations the authenticated user is a member of.
`leave_organization`	Destructive	Not offered	Remove the authenticated user from an organization.
`get_org_billing_information`	Read	Not offered	Retrieve the current billing details for an organization.
`get_org_billing_refill_information`	Read	Not offered	Retrieve the recharge amount for an organization’s billing.
`list_audit_logs`	Read	Not offered	Retrieve a paginated, filterable list of audit events for your organization.

Agent onboarding

Tool	Access	Sign-in	Description
`start_agent_onboarding`	Write	Not offered	Start the device-authorization flow a bot uses to get its own Bland API key, plus a phone number if its owner starts the Agent Phone Plan.
`poll_agent_onboarding`	Write	Not offered	Poll for the result of a device-authorization flow started with /v1/agent/onboarding/start.
`approve_agent_onboarding`	Destructive	Not offered	Claim a device-authorization code after the owner signs up in the browser.

Agent Phone Plan

Tool	Access	Sign-in	Description
`cancel_agent_phone_plan`	Destructive	Not offered	Cancel your organization’s Agent Phone Plan immediately, with an optional offboarding survey.
`submit_agent_phone_plan_cancel_feedback`	Write	Not offered	Answer the optional Agent Phone Plan offboarding survey after you’ve canceled.

Prompts (Legacy pathways)

Over an API key connection, clients that support MCP prompts (Claude Code surfaces them as slash commands) also get these prompts. Sign-in connections don’t get them. All three work with legacy pathways through search_bland_docs, get_bland_doc, bland_api_get, call_bland_api, create_call, and get_call_log. For Bland Agents, use the Norm plugin commands instead.

Prompt	Arguments	What it does
`bland_pathways`	None	Loads Bland’s pathway doctrine into the session: the node and edge model, node types, variable extraction, routing, the hard rules, and the create, test, and publish workflow.
`build_pathway`	`goal`	Plans and builds a new pathway for the goal: create, save a version, self-validate, and place one test call. Asks before publishing to production.
`edit_pathway`	`pathway_id`, optional `change`	Inspects and safely edits an existing pathway, preserving the nodes and edges you don’t change, then re-validates and tests. Asks what to change when `change` is omitted, and asks before publishing.

Docs for agents: llms.txt

​Dedicated tools

​Agents

​Calls

​Choose the call resource before invoking a tool

​Analytics

​Evals

​Knowledge and code

​Billing

​Documentation

​API passthrough

​Setup

​Pathways (Legacy)

​API Reference tools

​Agent inbound numbers

​Agent versions

​Agent checks

​Agent experiments

​Agent identity

​Agent dispositions

​Agent library

​Migration to agents

​Calls API

​Batches

​SMS

​Numbers

​Blocked numbers

​Custom dialing pools

​SIP trunks

​Custom Twilio accounts

​Knowledge bases

​Custom tools

​Contacts

​Memory

​Agent testing

​Evals API

​Guard rails

​Prompts library

​Widgets

​Live translation

​Triage

​Account

​API keys

​Organizations

​Agent onboarding

​Agent Phone Plan

​Prompts (Legacy pathways)

Dedicated tools

Agents

Calls

Choose the call resource before invoking a tool

Analytics

Evals

Knowledge and code

Billing

Documentation

API passthrough

Setup

Pathways (Legacy)

API Reference tools

Agent inbound numbers

Agent versions

Agent checks

Agent experiments

Agent identity

Agent dispositions

Agent library

Migration to agents

Calls API

Batches

SMS

Numbers

Blocked numbers

Custom dialing pools

SIP trunks

Custom Twilio accounts

Knowledge bases

Custom tools

Contacts

Memory

Agent testing

Evals API

Guard rails

Prompts library

Widgets

Live translation

Triage

Account

API keys

Organizations

Agent onboarding

Agent Phone Plan

Prompts (Legacy pathways)