Canvas Medical connector for AI agents

Tools your clinical agent reaches for on Canvas Medical, scoped per user.

CALL ANY TOOL

74 tools for clinical records in Canvas Medical over FHIR: read and search patients, encounters, conditions, medications, and observations, book appointments, add chart entries, send patient messages, and manage tasks.

canvasmedical_allergen_search

Search allergen

Search the Canvas (FDB) allergen catalog by text or code. Returns a FHIR Bundle of Allergen resources with id and FDB, RxNorm and SNOMED codings. Use it to get the FDB allergen code that canvasmedical_allergy_intolerance_create needs. Canvas requires text or code. If the result has a `next` link, use `canvasmedical_next_page` to get more.

Parameters

Name

Type

Required

Description

code

string

Optional

Allergen code as system|code, with system http://www.nlm.nih.gov/research/umls/rxnorm or http://snomed.info/sct, e.g. http://snomed.info/sct|373270004. Provide this or text.

count

integer

Optional

Maximum number of results per page. Canvas defaults to 10 and caps it at 100, e.g. 50.

text

string

Optional

Case-insensitive partial match on the allergen name, e.g. penicillin. Provide this or code.

canvasmedical_allergy_intolerance_create

Create allergy intolerance

canvasmedical_allergy_intolerance_mark_entered_in_error

Allergy intolerance mark entered in error

canvasmedical_allergy_intolerance_read

Read allergy intolerance

canvasmedical_appointment_search

Search appointment

canvasmedical_appointment_update

Update appointment

Update an appointment in place: reschedule it, change its status (for example cancel or check in), or change its type, reason or location. Returns an empty body on success. Use it to change an existing appointment; use canvasmedical_appointment_create for a new one. Send every required field, even unchanged ones; omitted optional fields keep their values. Run canvasmedical_appointment_read first to copy the current values.

Parameters

Name

Type

Required

Description

appointment_id

string

Required

Canvas ID of the appointment to update, e.g. 12345.

end

string

Required

End time in FHIR instant format, after start, e.g. 2024-06-03T14:00:00Z.

location_id

string

Required

Canvas ID of the practice location, from canvasmedical_location_search, e.g. 24680.

practitioner_id

string

Required

Canvas ID of the practitioner for the appointment, e.g. 67890. Every appointment needs one.

start

string

Required

Start time in FHIR instant format, e.g. 2024-06-03T13:30:00Z (YYYY-MM-DDTHH:MM is required).

status

string

Required

Appointment status: proposed (unconfirmed), pending (attempted), booked (confirmed), arrived, checked-in (roomed), fulfilled (exited), noshow or cancelled. One of: `proposed`, `pending`, `booked`, `arrived`, `checked-in`, `fulfilled`, `noshow`, `cancelled`.

appointment_type_code

string

Optional

Code of a schedulable appointment (note) type configured in Canvas, e.g. 448337001 (telemedicine). Omit to use the instance's default type.

appointment_type_system

string

Optional

Code system of appointment_type_code: http://snomed.info/sct (default) or INTERNAL. One of: `http://snomed.info/sct`, `INTERNAL`.

meeting_link

string

Optional

Video meeting URL for a telehealth appointment, e.g. https://video.example.com/room/123. Defaults to the practitioner's meeting room.

patient_id

string

Optional

Canvas ID of the patient, e.g. 12345. Leave empty only for appointment types without a patient. Canvas does not let an update change the patient; cancel and rebook instead.

reason_code

string

Optional

Code of a structured reason for visit configured in Canvas, e.g. INIV. Leave empty unless the instance uses structured reasons.

reason_text

string

Optional

Reason for visit text, e.g. Follow-up on blood pressure. With a structured reason code this becomes its comment.

canvasmedical_care_plan_read

Read care plan

canvasmedical_communication_create

Create communication

canvasmedical_communication_read

Read communication

canvasmedical_communication_search

Search communication

canvasmedical_coverage_search

Search coverage

canvasmedical_device_read

Read device

canvasmedical_document_reference_read

Read document reference

canvasmedical_document_reference_search

Search document reference

canvasmedical_goal_search

Search goal

canvasmedical_immunization_create

Create immunization

canvasmedical_immunization_mark_entered_in_error

Immunization mark entered in error

canvasmedical_immunization_read

Read immunization

canvasmedical_medication_read

Read medication

canvasmedical_medication_request_search

Search medication request

canvasmedical_medication_statement_search

Search medication statement

canvasmedical_observation_create

Create observation

Record one vital sign for a patient; Canvas files it in a Vitals command. Returns an empty body on success; Canvas puts the new resource's ID only in the Location response header. Use it for blood pressure, weight, height, pulse, temperature, oxygen saturation, respiration rate, waist circumference, pulse rhythm or a vitals note. Canvas accepts only vital signs here, not lab results. Without panel_id Canvas creates a new vitals panel; find an existing panel with canvasmedical_observation_search (code 85353-1).

Parameters

Name

Type

Required

Description

patient_id

string

Required

Canvas ID of the patient the vital sign is for, e.g. 12345.

vital_code

string

Required

LOINC code of the vital sign: 85354-9 blood pressure, 29463-7 weight, 8302-2 height, 8867-4 pulse rate, 8310-5 body temperature, 59408-5 oxygen saturation, 2708-6 arterial oxygen saturation, 9279-1 respiration rate, 56086-2 waist circumference, 8884-9 pulse rhythm, 80339-5 vitals note. One of: `85354-9`, `29463-7`, `8302-2`, `8867-4`, `8310-5`, `2708-6`, `59408-5`, `9279-1`, `56086-2`, `80339-5`, `8884-9`.

diastolic

number

Optional

Diastolic blood pressure in mmHg, for blood pressure (85354-9) only, e.g. 80.

effective_date_time

string

Optional

When the vital sign was taken, ISO 8601, e.g. 2024-06-01T10:30:00Z. Defaults to now.

panel_id

string

Optional

Canvas ID of an existing vital signs panel observation to add this reading to, e.g. 12345. Each panel holds one reading per vital type.

performer_id

string

Optional

Canvas ID of the practitioner who took the measurement, e.g. 67890.

systolic

number

Optional

Systolic blood pressure in mmHg, for blood pressure (85354-9) only, e.g. 120.

unit

string

Optional

Unit for value. Defaults: weight oz (also lb, kg), height in (also cm), pulse and respiration bpm, temperature °F, oxygen %, waist cm (also in). Other units return 422, e.g. lb.

value

number

Optional

Numeric value for weight, height, pulse rate, temperature, oxygen saturation, respiration rate or waist circumference, e.g. 160.

value_string

string

Optional

Text value for pulse rhythm (Regular, Irregularly Irregular, Regulary Irregular) or a vitals note. For blood pressure use systolic and diastolic instead.

canvasmedical_observation_read

Read observation

canvasmedical_patient_read

Read patient

canvasmedical_patient_search

Search patient

canvasmedical_procedure_search

Search procedure

canvasmedical_questionnaire_read

Read questionnaire

canvasmedical_related_person_read

Read related person

canvasmedical_related_person_search

Search related person

canvasmedical_service_request_search

Search service request

For more tools, view docs.

Build your Agent

Same auth pattern across LangChain, OpenAI, Anthropic, and Google ADK.

import { ScalekitClient } from "@scalekit-sdk/node";
import { createReactAgent } from "@langchain/langgraph/prebuilt";

const sk = new ScalekitClient(env.SCALEKIT_ENV_URL, env.SCALEKIT_CLIENT_ID, env.SCALEKIT_CLIENT_SECRET);

// Canvas Medical tools scoped to this user
const { tools } = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["canvasmedical"], toolNames: [
    "canvasmedical_patient_search",
    "canvasmedical_condition_search",
    "canvasmedical_appointment_create"] },
  pageSize: 100,
});

const agent = createReactAgent({ llm, tools });
await agent.invoke({ messages: [{ role: "user", content: "Find Maria Lopez, list her active conditions, and book her a follow-up next Tuesday" }] });

import OpenAI from "openai";
import { ScalekitClient } from "@scalekit-sdk/node";

const sk = new ScalekitClient(env.SCALEKIT_ENV_URL, env.SCALEKIT_CLIENT_ID, env.SCALEKIT_CLIENT_SECRET);
const openai = new OpenAI();

const { tools } = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["canvasmedical"] }, pageSize: 100,
});

const res = await openai.chat.completions.create({
  model: "gpt-5",
  messages: [{ role: "user", content: "Find Maria Lopez, list her active conditions, and book her a follow-up next Tuesday" }],
  tools,
});

// Execute the tool call with the user's vaulted Canvas Medical OAuth token
await sk.tools.executeTool(res.choices[0].message.tool_calls[0], "user_123");

import Anthropic from "@anthropic-ai/sdk";
import { ScalekitClient } from "@scalekit-sdk/node";

const sk = new ScalekitClient(env.SCALEKIT_ENV_URL, env.SCALEKIT_CLIENT_ID, env.SCALEKIT_CLIENT_SECRET);
const anthropic = new Anthropic();

const { tools } = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["canvasmedical"] }, pageSize: 100,
});

const msg = await anthropic.messages.create({
  model: "claude-sonnet-5",
  max_tokens: 1024,
  messages: [{ role: "user", content: "Find Maria Lopez, list her active conditions, and book her a follow-up next Tuesday" }],
  tools,
});

// Tool call runs with the user's vaulted Canvas Medical OAuth token
await sk.tools.executeTool(msg.content, "user_123");

import { Agent } from "@google/adk/agents";
import { ScalekitClient } from "@scalekit-sdk/node";

const sk = new ScalekitClient(env.SCALEKIT_ENV_URL, env.SCALEKIT_CLIENT_ID, env.SCALEKIT_CLIENT_SECRET);

const { tools } = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["canvasmedical"] }, pageSize: 100,
});

const agent = new Agent({
  name: "clinical_agent",
  model: "gemini-2.5-pro",
  instruction: "Canvas Medical tools scoped to this user",
  tools,
});

await agent.run("Find Maria Lopez, list her active conditions, and book her a follow-up next Tuesday");

Try these prompts

Copy any prompt into your agent. Each maps directly to a Canvas Medical tool. Click to copy, paste into your agent, done.

Patient records

Copy the prompt

Copied

Search for patients named Maria Lopez born in 1984.

Copy the prompt

Copied

List active conditions for this patient.

Copy the prompt

Copied

Show this patient's current medication list.

Scheduling

Copy the prompt

Copied

Find open slots on Dr. Chen's schedule next week.

Copy the prompt

Copied

Book a follow-up appointment for this patient with Dr. Chen on Tuesday at 10am.

Copy the prompt

Copied

Reschedule this appointment to Thursday afternoon.

Charting and tasks

Copy the prompt

Copied

Add a penicillin allergy to this patient's chart.

Copy the prompt

Copied

Record a blood pressure reading of 128/82 for this patient.

Copy the prompt

Copied

Create a task for the front desk to verify this patient's insurance.

SEE HOW AUTH WORKS

Each user signs in to Canvas Medical once, and Scalekit stores and refreshes their tokens. Credentials stay vaulted, every call is scope checked, and every action is logged.

Authorize

Your user connects

Canvas Medical

once. We tie it to their identity and the meetings they approved — no shared bot account, no org-wide access

Who:

user ‘A’

when:

Once per user

access:

Limited to user

Store

Their

Canvas Medical

token lives in a vault scoped to them. User A's meetings are never reachable by an agent acting for user B, even on the same connection

vault:

encrypted

scope:

per-user

tokens:

auto-refreshed

Resolve

When your agent calls a

Canvas Medical

tool, we fetch the right token server-side. It never touches your agent, never appears in the LLM context, never shows up in your logs

speed:

~40ms

check:

before every call

seen by:

nobody

Audit

Every

Canvas Medical

tool call is logged — who triggered it, which meeting was fetched, what came back. 90 days of history, tied to the user who authorized it

history:

90 days

export:

SIEM-ready

logged:

every call

Test other agents

See the same per-user auth pattern across other healthcare connectors.

Support and Ops Teams

Meeting prep

Pulls agenda, participant context, and open action items before every meeting.

Engineering Teams

Slack triage

Polls Slack for new messages, classifies bugs and support requests with a LangGraph router, files GitHub issues or Zendesk tickets, and confirms in the thread.

Why Scalekit

Secure your agent's access. Connectors ship in minutes

01.

Shared tokens break per-user analytics

A shared Canvas Medical token looks fine in a demo. In production every chart entry looks like one service account, and you cannot tell which user triggered it. Scalekit resolves the credential of the actual user who triggered the agent, never a shared bot.

// shared token
audit → bot_service_account

// scalekit
audit → user_abc ✓

02.

Authentication is not authorization

03.

Multi-tenancy is architectural

04.

Canvas Medical today. Ten connectors tomorrow.

“Our agents act across Salesforce, Gong, Google Drive, and more, on behalf of every customer. Scalekit behind the scenes meant we can keep adding tools without ever rebuilding how credentials or tool calling work.”

Venu Madhav Kattagoni

Head of Engineering / Von

FAQs

Frequently Asked Questions

Does the agent access Canvas Medical as the user or through a shared service account?

As the user. Each user signs in to Canvas Medical once with SMART on FHIR, and Scalekit stores and refreshes their tokens. Every chart read and write in your audit trail is attributed to that user, not a shared service account.

Where is the Canvas Medical OAuth token stored?

In Scalekit's managed AES-256 token vault, namespaced per tenant. Refresh is automatic. Revocation is a single dashboard action. Credentials never appear in prompts, logs, or LLM context.

Can I limit what the agent does in Canvas Medical?

Yes. Filter by tool name in listScopedTools to expose only what you want, for example the read and search tools without canvasmedical_patient_update or canvasmedical_communication_create. The connector has no delete tools. Scalekit also enforces scope checks before every API call.

What happens when a user revokes Canvas Medical access?

The connection is invalidated on the next tool call. Subsequent requests for that user fail closed with a clear error. Other users in the tenant remain unaffected. The event is logged for audit.

How does the agent correct a chart entry in Canvas Medical?

With the mark entered in error tools, such as canvasmedical_condition_mark_entered_in_error and canvasmedical_allergy_intolerance_mark_entered_in_error. They flag the entry as recorded by mistake instead of deleting it.