mcpbeat Sign in

Matlab Connect Ble Agent Skill

> Discover and connect to Bluetooth Low Energy (BLE) peripheral devices from MATLAB. Use this skill when the user wants to scan for BLE devices, connect to a BLE peripheral, read or write BLE characteristics, subscribe to notifications, or any task that requires a BLE connection as a prerequisite (e.g., read sensor data over BLE, monitor heart rate, log accelerometer, Bluetooth Low Energy, blelist, ble device, BLE characteristic, GATT, service UUID, scan for devices, BLE sensor, subscribe notify, BLE read, BLE write, peripheral device, BLE connect, wireless sensor.

4k tokens
context cost
the whole folder, loaded on every use
3
files
ships runnable scripts
0
copies elsewhere
how many repositories repackaged it
865
stars on the repo
on the repository, not the skill itself

Install

one command, takes just this skill from the repository
npx skills add https://github.com/matlab/matlab-agentic-toolkit --skill matlab-connect-ble

The instruction itself

20 sections, as written by the author

Connect to BLE Device in MATLAB

Guides the agent through discovering, connecting to, and accessing

characteristics on BLE peripheral devices, ensuring efficient use of

filtering APIs and correct function signatures.

BLE functions are part of MATLAB — no additional toolbox is required.

Do not reference the Bluetooth Toolbox for BLE functionality — it is

not needed.

When to Use

  • User wants to scan for or connect to a BLE device
  • User wants to read, write, or subscribe to BLE characteristics
  • User's task requires a BLE connection as a prerequisite (sensor data,

IoT communication, wearable monitoring)

  • User mentions BLE, Bluetooth Low Energy, GATT, service UUID, or

characteristic UUID

When NOT to Use

  • Bluetooth Classic connections (bluetooth() object — different API)
  • BLE peripheral mode (MATLAB acting as a BLE server/advertiser)
  • Arduino BLE via Arduino Support Package (see matlab-connect-arduino)
  • Support package installation or Bluetooth adapter driver issues
  • Simulink BLE blocks

Workflow

Follow these steps in order. Do not skip steps or guess values.

Step 0: Check workspace for existing connections

Always check first — a connected BLE device stops advertising, so

scanning will never find it:

% Check for existing ble objects in workspace
bleVars = whos;
bleVars = bleVars(strcmp({bleVars.class}, 'ble'));

If a ble object exists and its Connected property is 1:

  • Tell the user: "Found existing connection to [Name] on [Address]."
  • Ask if they want to reuse it or connect to a different device.
  • STOP and wait for user response.
  • If reusing, skip to Step 4.

If a ble object exists but Connected is 0, it is stale — proceed

to Step 1.

Step 1: Discover devices

Use filtered scanning when possible. **Do not scan all devices and then

search through results manually.**

| User provides | Use this call |

|---------------|---------------|

| Device name | blelist("Name", "<name>") |

| Service UUID | blelist("Services", "<uuid>") |

| Nothing specific | blelist("Timeout", 10) |

% Filter by name (prefix match)
devices = blelist("Name", "HeartSensor");

% Filter by advertised service UUID
devices = blelist("Services", "180D");

% General scan with extended timeout
devices = blelist("Timeout", 10);

Filters can be combined:

devices = blelist("Name", "Sensor", "Timeout", 15);

blelist returns a table with columns: Name, Address, RSSI, Advertisement.

Results are sorted by signal strength (strongest first).

Present results to the user.

Step 1b: Handle "no device found"

If blelist returns an empty table:

  • Check Step 0 again — the device may already be connected (and

therefore not advertising).

  • Ask the user: "No BLE device was found. Is the device powered on

and advertising?"

  • Suggest common fixes:
  • Ensure the device is in advertising/pairing mode
  • Move closer (BLE range is typically 10-30 meters)
  • Check that no other application holds the connection (nRF Connect,

LightBlue, another MATLAB session)

  • Try a longer timeout: blelist("Timeout", 20)
  • Restart the device to force re-advertising
  • STOP and wait. Do not proceed until the user confirms the device

is available.

Step 2: Let user choose device

Present the discovered devices and ask which to connect to.

Never auto-select a device.

If only one device matches a filtered scan (e.g., blelist("Name","MyDevice")

returned exactly one row), confirm with the user: "Found [Name] at

[Address]. Shall I connect?"

Step 3: Connect

% Connect by name (substitute the user's actual device name)
b = ble("HeartSensor");

% Connect by address (use when name is empty or ambiguous)
b = ble("A1B2C3D4E5F6");

After connecting, confirm to the user:

  • Device name and address
  • Number of services and characteristics available

If the connection succeeds but a service shows "Access denied":

  • Most likely cause: Another MATLAB session (or application) holds a

connection to the same device. The Windows BLE stack denies access to

services that are already claimed.

  • Fix: Ask the user to disconnect from all other sessions (clear b

in the other session, or close it), then reconnect from this session.

  • Do NOT diagnose this as a pairing/bonding issue — that is rarely

the cause for custom services that previously worked.

If the connection fails entirely:

  • Report the exact error to the user
  • Common causes: device out of range, device already connected elsewhere,

Bluetooth adapter disabled

  • Do not retry automatically — wait for user instruction

Step 4: Access characteristics

Inspect available services and characteristics:

% View all services
b.Services

% View all characteristics (includes service, name, UUID, attributes)
b.Characteristics

To access a specific characteristic, use the two-UUID signature:

% Access by service UUID + characteristic UUID
c = characteristic(b, "<serviceUUID>", "<characteristicUUID>");

% Access by service name + characteristic name (standard services only)
c = characteristic(b, "Battery Service", "Battery Level");

Important: characteristic() always requires both a service

identifier and a characteristic identifier. Never call it with just

one UUID.

**Never invent or assume device names, addresses, service UUIDs, or

characteristic UUIDs.** Use only values provided by the user or

discovered from blelist / b.Characteristics.

If characteristic access fails:

  • "No characteristic found" → UUID is wrong. Inspect b.Characteristics

to find the correct service and characteristic UUIDs.

  • "Access denied" → another session holds the connection (same fix as

Step 3: disconnect from other sessions first).

  • Never guess UUIDs — ask the user or read them from the device.

Step 5: Read / Write / Subscribe

% Read raw bytes from a characteristic (returns numeric array)
data = read(c);

% Read with timestamp
[data, timestamp] = read(c);

% Read oldest buffered notification (use in DataAvailableFcn callbacks)
data = read(c, "oldest");

% Write data to a characteristic
write(c, [1 0 1 0]);

% Write without response (faster, no confirmation)
write(c, [1 0 1 0], "WithoutResponse");

% Subscribe to notifications
subscribe(c);
c.DataAvailableFcn = @(src, evt) disp(read(src));

% Unsubscribe
unsubscribe(c);

**read and write operate on characteristic objects, not on the ble

object directly.** Never use read(b, uuid) — this will error.

Key Functions

| Function | Purpose |

|----------|---------|

| blelist | Scan for BLE peripherals (with optional Name/Services/Timeout filters) |

| ble(name) | Connect to a BLE device by advertised name |

| ble(address) | Connect to a BLE device by MAC address |

| characteristic(b, svcID, charID) | Access a characteristic (requires both service and characteristic identifiers) |

| read(c) | Read data from a characteristic object |

| write(c, data) | Write data to a characteristic object |

| subscribe(c) | Subscribe to notifications/indications |

| unsubscribe(c) | Unsubscribe from notifications/indications |

All functions are part of MATLAB — no toolbox required.

Patterns

Single device by name

% TEMPLATE — not executable without hardware
% Step 0: Check workspace
bleVars = whos;
bleVars = bleVars(strcmp({bleVars.class}, 'ble'));

% Step 1: Filtered discovery
devices = blelist("Name", "TempLogger");

% Step 3: Connect
b = ble("TempLogger");

% Step 4-5: Access and read custom characteristic
c = characteristic(b, "181A", "2A6E");
data = read(c);

Find device by service UUID

% TEMPLATE — not executable without hardware
% Find devices advertising a specific service
devices = blelist("Services", "180D");

% Connect to the first match (after user confirms)
b = ble(devices.Name(1));

% List characteristics under that service
b.Characteristics

Reuse existing connection

% TEMPLATE — not executable without hardware
% Check if 'b' already exists and is connected
if exist('b', 'var') && isa(b, 'ble') && b.Connected
    disp("Reusing existing connection to " + b.Name);
else
    b = ble("MyDevice");
end

% Access characteristic on existing connection
c = characteristic(b, "180F", "2A19");
batteryLevel = read(c);

Subscribe to notifications

% TEMPLATE — not executable without hardware
c = characteristic(b, "180D", "2A37");
subscribe(c);
c.DataAvailableFcn = @(src, evt) processHeartRate(read(src));

% Later: clean up
unsubscribe(c);

Log characteristic data via notifications

When the user wants to log or collect BLE data over a time period,

use DataAvailableFcn with subscribe/unsubscribe — **do not poll with

read(c) in a loop** (polling misses data between reads and may flush the

buffer).

Deploy scripts/bleDataCollector.m to the user's working directory and run:

% TEMPLATE — not executable without hardware
% Log notifications to a tab-separated text file for 10 seconds:
bleDataCollector(b, "<serviceUUID>", "<charUUID>", 10, "sensorLog.txt")

Output file format (one row per notification):

2026-06-25 10:30:01.0529	3 12 255
2026-06-25 10:30:01.2478	4 13 0

Key points for notification-based logging:

  • Set DataAvailableFcn to a callback that calls read(src, "oldest")

using "oldest" reads buffered notification data in order; "latest" may

flush and lose earlier packets

  • Call subscribe(c) to start receiving notifications, unsubscribe(c) to

stop

  • pause(duration) keeps MATLAB alive while callbacks fire — this is NOT

polling, it is the idle loop that allows notifications to be delivered

  • Always unsubscribe, clear DataAvailableFcn, and fclose when done
  • Log to .txt with tab separation — BLE packets vary in length, so

fixed-column CSV is impractical

Script: scripts/bleDataCollector.m (bundled with this skill — deploy to the user's working directory before use)

Common Mistakes

| Mistake | Why It's Wrong | Correct Approach |

|---------|---------------|-----------------|

| Using unfiltered blelist when name/service is known | Scans all devices unnecessarily; slower and noisier | Use blelist("Name","...") or blelist("Services","...") |

| Calling read(b, uuid) on the ble object | read works on characteristic objects, not ble objects — will error | Create characteristic first: c = characteristic(b, svc, char), then read(c) |

| Using blescanner | This function does not exist in MATLAB | Use blelist (with optional name-value filters) |

| Saying BLE requires "Bluetooth Toolbox" | BLE functions are part of MATLAB — no toolbox needed | Do not check for or reference the Bluetooth Toolbox |

| Scanning for a device that is already connected | Connected devices stop advertising — scan will never find them | Check workspace for existing ble objects first (Step 0) |

| Calling characteristic(b, uuid) with one UUID | Requires both service and characteristic identifiers | Use characteristic(b, serviceUUID, characteristicUUID) |

| Auto-selecting a device without user confirmation | User may have multiple devices; wrong choice wastes time | Always present options and ask which device to connect to |

| Diagnosing "Access denied" as a pairing issue | Usually caused by another session holding the connection, not missing pairing | Ask user to disconnect from other MATLAB sessions first |

Conventions

  • Always check workspace for existing ble connections before scanning — connected devices stop advertising
  • Use filtered blelist when name or service UUID is known — never scan all and search manually
  • Always present discovered devices to the user before connecting
  • Never auto-select a device — let the user choose
  • read/write/subscribe operate on characteristic objects, not ble objects
  • characteristic() requires two identifiers: service + characteristic
  • BLE is part of MATLAB — never state or check for toolbox requirements
  • BLE connections are exclusive — only one application can hold a connection to a given device at a time

----

Copyright 2026 The MathWorks, Inc.

----

Other skills for the same job

different authors, same section of the catalogue
Protocolsio Integration
by christophacham
×4

Integration with protocols.io API for managing scientific protocols. This skill should be used when working with protocols.io to search, create, update, or publish protocols; manage protocol steps and materials; handle discussions and comments; organize workspaces; upload and manage files; or integrate protocols.io functionality into workflows. Applicable for protocol discovery, collaborative protocol development, experiment tracking, lab protocol management, and scientific documentation.

16k tokens
Tailored Resume Generator
by frostant
×4

Analyzes job descriptions and generates tailored resumes that highlight relevant experience, skills, and achievements to maximize interview chances

3k tokens
Excalidraw Diagram Generator
by github
vendor ×3

Generate Excalidraw diagrams from natural language descriptions. Use when asked to "create a diagram", "make a flowchart", "visualize a process", "draw a system architecture", "create a mind map", or "generate an Excalidraw file". Supports flowcharts, relationship diagrams, mind maps, and system architecture diagrams. Outputs .excalidraw JSON files that can be opened directly in Excalidraw.

36k tokens scripts
Expo Dev Client
by openai
vendor ×3

Build and distribute Expo development clients locally or via TestFlight

961 tokens
Executing Plans
by ZhanlinCui
×3

Use when you have a written implementation plan to execute in a separate session with review checkpoints

542 tokens
Anndata
by christophacham
×3

Data structure for annotated matrices in single-cell analysis. Use when working with .h5ad files or integrating with the scverse ecosystem. This is the data format skill—for analysis workflows use scanpy; for probabilistic models use scvi-tools; for population-scale queries use cellxgene-census.

16k tokens
Benchling Integration
by christophacham
×3

Benchling R&D platform integration. Access registry (DNA, proteins), inventory, ELN entries, workflows via API, build Benchling Apps, query Data Warehouse, for lab data management automation.

14k tokens
Biopython
by christophacham
×3

Comprehensive molecular biology toolkit. Use for sequence manipulation, file parsing (FASTA/GenBank/PDB), phylogenetics, and programmatic NCBI/PubMed access (Bio.Entrez). Best for batch processing, custom bioinformatics pipelines, BLAST automation. For quick lookups use gget; for multi-service integration use bioservices.

24k tokens

How to use it

Copy the folder

Take matlab/matlab-connect-ble from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.