Index of every page on this site: llms.txt
All skills
aws avatar

/amazon-braket

@12d7f78
Run it yourself

Runs quantum workflows on AWS / Amazon Braket: circuits and algorithms, QPU/device and ARN discovery, LocalSimulator, tasks, program sets, Hybrid Jobs, S3 results, pricing, account setup, and spending limits. Applies to quantum programs and execution, AHS, OpenQASM, pulse control, reservations, notebooks, or Qiskit or PennyLane integrations.

Use this Skill: https://skilld.dev/gh/aws/agent-toolkit-for-aws/amazon-braket

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈90 tokens for metadata: the name and description. ≈4.5k when used: this file. ≈16k more on demand in 12 files.

Description uses 4.5% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈359 characters, or 4.5%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

<!-- markdownlint-disable MD013 -->

Amazon Braket

Primitives

The vocabulary of a Braket workflow, and which reference to open for each.

Primitive What it is Related References Open it when the request involves
Device A simulator or QPU, identified by a region-scoped ARN devices.md anything about a device: what exists, discovering or filtering the fleet, availability and status (online, offline, retired), which region a device lives in, ARNs, choosing a device for a workload, qubit count, connectivity or topology, native gates, fidelities, calibration data, queue depth, shot and gate limits, paradigm (gate-model vs analog Hamiltonian simulation), whether a device supports program sets, simulators and local emulators
Program The workload/input — one executable (Circuit, AHS, OpenQASM). Which type is legal depends on the device's paradigm - -
Quantum task One program + shots, run once (the atomic unit Braket meters) - -
Task batch Many independent tasks — SDK-only fallback for when a program set does not fit; works on all devices program-sets.md see the Program set row; also running multiple programs
Program set Many programs in one service-side task — preferred way to run multiple programs instead of task batch program-sets.md running more than one program: parameter sweeps, scanning parameter values, task batches, run_batch, several circuits submitted together, attaching observables across programs, and minimizing per-task fees when many programs run, program sets. Also getting started with program sets
Hybrid job Managed classical-quantum loop that orchestrates many tasks hybrid-job.md hybrid jobs: @hybrid_job, algorithm scripts and source modules, entry_point, embedded simulators, BYOC and custom container images, CUDA-Q, job execution roles, hyperparameters, checkpoints, and retrieving job results
Spending limit Service-side hard cap that rejects QPU tasks — the only true enforcement (not SDK) spending-limit.md capping or enforcing spend: spending limits (create, update, delete, search), and cost guardrails
Cost tracking In-session cost estimate (not enforcement) spending-limit.md in-session cost tracking with Tracker
Pricing Current Braket rates and pricing models pricing.md current price or rate, simulator pricing, QPU per-task/per-shot pricing, Hybrid Job or reservation pricing, AWS Pricing API, bulk price list
Reservation Exclusive device access for a booked window (Braket Direct) reservation.md reservations: DirectReservation, reservation_arn, booking/cancelling a reservation, running tasks or hybrid jobs during a reservation window
Notebooks Jupyter/JupyterLab notebook instances for exploring Braket notebooks.md creating or managing a Braket notebook instance, Jupyter, getting-started environment for Braket
Pulse control / gate calibrations Native gate calibrations and custom pulse sequences on supported QPUs pulse-control.md pulse control, gate calibrations, gate_calibrations, refresh_gate_calibrations, gate_definitions
Experimental capabilities Opt-in device features outside the published specifications experimental-capabilities.md features that not all QPUs support, EnableExperimentalCapability, experimental_capabilities
Library integrations Third-party frameworks using Braket-owned providers or plugins Qiskit, PennyLane Qiskit or PennyLane workflows targeting Braket; provider/plugin behavior and Braket-specific compilation, submission, or result retrieval

Each reference carries the domain detail for its own area — field paths, key names, API shapes, and billing models.

Additional notes:

  • Setup — first-time account setup, prerequisites, roles, notebooks, and the results S3 bucket are covered in setup.md.

Critical Rules

These rules apply to every Braket request, whatever it involves.

  1. Use credentials with IAM-administrative permissions only for setup actions, such as creating IAM roles and policies, creating or updating spending limits, and creating S3 buckets or other resources. After setup, use workload credentials with permissions scoped to the workflow.

  2. The Amazon Braket Python SDK (pip install amazon-braket-sdk, imported as braket) is the primary entry point. Prefer it for every operation including Braket API operations, and understand what it covers by reading the docs or inspecting the SDK's modules locally.

    • Use local execution tools for running the Braket SDK, such as shell with python3 -c "<code>".
  3. The AWS MCP server is recommended for executing any other AWS API calls in this skill, especially operations not present in the Python SDK, although not required. Note: the AWS MCP's run_script tool executes code in a minimal sandbox without Braket libraries, so prefer using other tools for code execution, especially when using the Braket SDK.

  4. A small set of primitives composes every workflow — see Primitives for the vocabulary and the reference for each.

  5. Devices, quantum tasks, and hybrid jobs are region-scoped, so fan out across every Braket region whenever you use the API, CLI, or boto3 to search for resources. The SDK handles fanout for you where it can — AwsDevice.get_devices searches QPUs in all regions. Get the list of regions Braket supports from aws___get_regional_availability when the AWS MCP server is available, or from the supported devices and regions documentation. Resource ARNs containing a region may only be queried from that same region, otherwise you will see a ResourceNotFoundException.

  6. Open the matching reference before you write code or answer. Use the Primitives table to find and read references. Note: A request that asks for several things, e.g. executing a series of circuits and controlling the cost thereof, may require reading multiple reference files.

  7. Verify, never recall. Device ARNs and statuses, API signatures, supported features, and prices (and other values) all change and may post-date training data.

  8. Confirm a signature from documentation before you write code that calls it. This applies to the Braket SDK and to every third-party library that integrates with Braket. A guessed parameter or a guessed unit can produce a quantum task that runs, charges the customer, and returns a wrong answer.

    • For the Braket SDK, read the SDK reference docs.
    • For a third-party library, open its matching library reference from the Primitives table and verify signatures against the linked provider or plugin documentation.

    If no available tool reaches the documentation, read the signature from the installed package with shell:

    # subpackages: ahs, circuits, pulse, program_sets, ...
    PAGER=cat python -c "import braket; help(braket)"
    # names: DrivingField, AtomArrangement, ...
    python -c "import braket.ahs; print(dir(braket.ahs))"
    python -c "import inspect; from braket.ahs import DrivingField; \
    print(inspect.signature(DrivingField.from_lists)); print(inspect.getdoc(DrivingField.from_lists))"
  9. Attribute skill usage. In every Python script or Jupyter notebook that uses the Braket SDK, insert this block before any Braket SDK code, including import braket. In a notebook, run the cell after you add it, so that attribution starts before any Braket SDK code runs:

    # Amazon Braket agent skill: adds a usage tag to AWS requests. No other data is sent.
    import os
    _app = os.environ.get("AWS_SDK_UA_APP_ID", "")
    if "AWSSkill-Braket" not in _app:
        os.environ["AWS_SDK_UA_APP_ID"] = f"{_app}+AWSSkill-Braket#1.0.0".lstrip("+")

    If using the AWS CLI, append this skill's identifier to AWS_SDK_UA_APP_ID. The variable holds one value and other tools also write it, so always keep the existing value and add the identifier only once:

    # sh, bash, zsh, ksh
    case "${AWS_SDK_UA_APP_ID:-}" in
      *AWSSkill-Braket*) ;;
      *) export AWS_SDK_UA_APP_ID="${AWS_SDK_UA_APP_ID:+$AWS_SDK_UA_APP_ID+}AWSSkill-Braket#1.0.0" ;;
    esac
    # PowerShell — session-scoped
    if ($env:AWS_SDK_UA_APP_ID -notlike '*AWSSkill-Braket*') {
        $env:AWS_SDK_UA_APP_ID = (@($env:AWS_SDK_UA_APP_ID, 'AWSSkill-Braket#1.0.0') -ne '' -ne $null) -join '+'
    }

    This marks AWS calls as originating from use with this skill. This is not needed when using AWS MCP tools as the AWS MCP automatically attributes during tool calls.

  10. Before the first QPU task on each device, search for an existing spending limit on that device ARN in its region; if none exists, create one with an amount the user confirms. Do not raise or delete a limit without an explicit user instruction. See spending-limit.md.

Guardrail — where this skill's own files live (MCP vs local install)

This skill can be loaded two ways, and they resolve the skill's own bundled files from different places. Determine how the skill was loaded before reading a reference:

  • Loaded through the AWS MCP retrieve_skill tool: The skill is not installed on the local filesystem. You MUST fetch each reference via retrieve_skill with the file parameter (e.g. file="references/devices.md"). Do NOT file_read these paths locally — they do not exist on disk.
  • Installed locally (e.g. ~/.kiro/skills/amazon-braket/, .kiro/skills/amazon-braket/, or ~/.claude/skills/amazon-braket/): Read files from the local skill directory using relative paths.

references/ is a sibling of this SKILL.md — resolve reference paths against that directory, not your working directory, and do not search the filesystem for them. If a skill tool returns this overview instead of the file you asked for, it did not fetch it: read it from that directory instead, and do not write code from memory because a reference read failed. This distinction applies only to the skill's own packaged files. User data and session artifacts are always read from and written to the user's working directory — do not cd before running a script that writes a relative artifact path.

Common mistakes

Symptom Cause Fix
Treats "task", "batch", "job" as interchangeable Primitive confusion Task = one run; batch = many parallel tasks; job = managed loop
Missing required parameter: filters on SearchQuantumTasks, SearchJobs, or SearchDevices filters is required on all three — only SearchSpendingLimits lets you omit it Pass filters=[] (--filters '[]') for an unfiltered search. A populated filter needs name and values, plus operator on tasks and jobs; SearchDevices has no operator member
Missing required parameter: clientToken when using AWS MCP run_script tool CreateQuantumTask, CreateJob, CancelQuantumTask, CreateSpendingLimit, UpdateSpendingLimit require a clientToken for idempotency. The SDK, CLI, and boto3 generate one automatically; the run_script tool requires explicit passing Pass clientToken=str(uuid.uuid4()) on APIs taking clientToken when calling through run_script. Otherwise, prefer the SDK for these operations when possible.
Builds program IR as JAQCD JAQCD is deprecated on Amazon Braket Use OpenQASM — see OpenQASM on Braket
Denies a feature or SDK construct exists Training data outdated Verify against the docs or GetDevice before saying it does not exist
Invents a class, method, or parameter Writing API names from memory Confirm the signature first. If uncertain, say so rather than inventing

Security considerations

Braket workflows touch IAM, S3, and (for hybrid jobs) container execution.

For details, see the Amazon Braket security documentation.

Reference links

Authoritative sources — prefer these over recalled details, since device ARNs, quotas, prices, and supported features change.

Official documentation (stable entry points — navigate/search from here)

For anything not covered here, search the documentation rather than guessing page slugs or recalling details: if the AWS MCP server is available, its aws___search_documentation tool can help; otherwise start from the Developer Guide or API Reference above and navigate.

GitHub

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

1 alert1d3 checks · Risk SAFE
  • Gen Agent Trust Hub1d

    This skill provides a robust interface for managing quantum computing workflows. It incorporates security considerations typical for developer-focused cloud tools, such as the use of local shell execution for SDK operations and the configuration of execution roles. The skill also handles data from external sources like task results and managed job logs, which is consistent with its intended use in processing quantum results. Users are encouraged to review generated code and policy templates to verify they match the desired security posture.

  • Socket1d

    No alerts

  • Snyk1d

    Risk: CRITICAL · 2 issues

Signed by skilld at 12d7f78. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 2 days ago
metadata
{
  "version": "3"
}

README badge

README badge for aws/agent-toolkit-for-aws/amazon-braket