Skip to main content
Pre-release

This release is a pre-release and may not be stable for production use.

Swarmauri Logo

PyPI - Downloads Hits PyPI - Python Version PyPI - License PyPI - swarmauri_standard Discord

Swarmauri Standard

swarmauri_standard provides first-party, ready-to-use Swarmauri components built on swarmauri_core interfaces and swarmauri_base component behavior. It is the standard component bundle for agents, chains, chunkers, conversations, documents, tools, toolkits, parsers, prompts, schema converters, metrics, similarities, deprecated distance compatibility shims, vectors, vector stores, tracing utilities, STT/TTS/VLM adapters, signing helpers, and key-provider components.

Why Swarmauri Standard?

swarmauri_standard provides maintained first-party components that users can install and instantiate directly. It also demonstrates the expected Swarmauri component style: typed base classes, direct class imports, dynamic serialization, and namespace-ready registration.

FAQ

Q: What kinds of components are included?

A: The package includes agents, chains, chunkers, conversations, documents, tools, toolkits, parsers, prompts, schema converters, metrics, similarities, deprecated distance compatibility shims, vectors, vector stores, tracing utilities, STT/TTS/VLM adapters, signing helpers, and key-provider components.

Q: Is this package the same as swarmauri?

A: No. swarmauri provides the namespace importer and plugin discovery layer. swarmauri_standard provides concrete first-party implementation classes.

Q: Should examples instantiate plugins directly?

A: Yes. Direct class imports and direct instantiation are the preferred documented workflow unless a package explicitly requires another integration pattern.

Features

  • Standard agents including question-answering, RAG, simple conversation, and tool-using agent patterns.
  • Tool and toolkit components such as calculator, HTTP request, readability, code extraction, temperature conversion, weather, and accessibility utilities.
  • Parser and document components for CSV, XML, OpenAPI specs, markdown-to-HTML, URLs, regular expressions, Python code, and phone-number extraction.
  • Chunkers, conversations, prompts, prompt templates, chains, pipelines, service registries, state objects, and tracing helpers for application workflows.
  • Deprecated distance compatibility shims, vectors, vector stores, measurements, metrics, norms, similarities, and related numerical components.
  • Model adapter classes for LLM, tool LLM, embedding, image generation, STT, TTS, OCR, and VLM workflows where provider configuration is available.
  • Signing and key-provider components that align with Swarmauri's security-oriented interfaces.
  • Dynamic JSON, YAML, and TOML serialization through ComponentBase, DynamicBase, and SubclassUnion.
  • Python 3.10, 3.11, 3.12, 3.13, and 3.14 support.

Installation

Install with uv:

uv add swarmauri_standard

Install with pip:

pip install swarmauri_standard

Usage

Use standard components directly from their implementation modules:

from swarmauri_standard.tools.CalculatorTool import CalculatorTool

calculator = CalculatorTool()
result = calculator("add", 5, 3)

assert result == {"operation": "add", "calculated_result": "8.0"}

Create documents and serialize them as Swarmauri components:

from swarmauri_standard.documents.Document import Document

doc = Document(content="Composable intelligence infrastructure")
payload = doc.model_dump_json()
restored = Document.model_validate_json(payload)

assert restored.type == "Document"
assert restored.content == doc.content

Hydrate registered standard components through a base-family union:

from pydantic import BaseModel

from swarmauri_base.DynamicBase import SubclassUnion
from swarmauri_base.toolkits.ToolkitBase import ToolkitBase
from swarmauri_standard.toolkits.Toolkit import Toolkit


class ToolkitEnvelope(BaseModel):
    toolkit: SubclassUnion[ToolkitBase]


payload = ToolkitEnvelope(toolkit=Toolkit()).model_dump_json()
restored = ToolkitEnvelope.model_validate_json(payload)

assert isinstance(restored.toolkit, Toolkit)

Component Families

The package includes these standard component families:

  • Agents: QAAgent, RagAgent, SimpleConversationAgent, and ToolAgent.
  • Chains and workflow: callable chains, chain steps, context chains, prompt context chains, pipelines, service registries, swarms, and round-robin task management.
  • Text and document processing: chunkers, documents, conversations, prompts, prompt templates, parsers, messages, and schema converters.
  • Tools and toolkits: calculator, arithmetic, HTTP requests, JSON requests, code extraction, code interpretation, readability scoring, weather, temperature conversion, parameters, accessibility toolkit, and generic toolkit support.
  • Vector and numerical work: vectors, TF-IDF vector store, deprecated distance compatibility shims, measurements, metrics, norms, similarities, pseudometrics, seminorms, matrices, tensors, and inner products.
  • AI model surfaces: LLMs, tool LLMs, embeddings, image generators, OCR, STT, TTS, and VLM classes for supported providers.
  • Runtime support: decorators, tracing, transports, state, logging, and utility classes.
  • Trust and security surfaces: signing and key-provider components that align with Swarmauri's core interfaces.

Distance Migration

The swarmauri_standard.distances module remains available only as a deprecated compatibility surface. Prefer the active metric and similarity families instead:

  • CosineDistance -> CosineSimilarity
  • EuclideanDistance -> EuclideanMetric
  • LevenshteinDistance -> LevenshteinMetric
  • JaccardIndexDistance -> JaccardIndexSimilarity
  • SorensenDiceDistance -> DiceSimilarity
  • ChebyshevDistance -> SupremumMetric
  • ManhattanDistance -> LpMetric(p=1)
  • MinkowskiDistance -> LpMetric(p=<order>)
  • SquaredEuclideanDistance -> EuclideanMetric or GaussianRBFSimilarity, depending on whether you need metric or similarity semantics
  • CanberraDistance, ChiSquaredDistance, and HaversineDistance do not yet have active drop-in replacements in the workspace

Foundational packages:

  • swarmauri provides namespace imports and plugin discovery.
  • swarmauri_core provides the interfaces implemented by these standard components.
  • swarmauri_base provides component base classes, mixins, and serialization behavior.
  • swarmauri_typing provides dynamic typing utilities used by the component model.

Related component packages:

When To Use This Package

Use swarmauri_standard when you want maintained first-party implementations that are ready for examples, prototypes, tests, and production workflows where the included component behavior fits. Use smaller standalone component packages when you want a narrower dependency surface for a specific provider, algorithm, or runtime integration.

License

Apache-2.0

Contributing

When adding standard components, inherit from the matching swarmauri_base class, satisfy the swarmauri_core interface, register the component type, document direct instantiation, and follow the Swarmauri SDK contribution guide.

Resumable tool calls

Pass either existing OpenAI or Gemini tool-model class to swarmauri_agent_runner.AgentRunner with a run store and an approval policy. The runner dispatches tools and the adapter preserves the provider transcript.

Standalone alpha TTS providers

OpenAI speech now lives in swarmauri_tts_openai; the bundled OpenaiTTS and legacy OpenAIAudioTTS imports emit deprecation warnings and delegate to it. DeepInfra and Together have separate swarmauri_tts_deepinfra and swarmauri_tts_together alpha packages. Live provider verification is pending; see verification tracking.

TTS lifecycle and canonical imports

New speech integrations use the standalone alpha provider packages, with verification issues in docs/verification/tts-providers.md. The bundled OpenaiTTS and LLM-named OpenAIAudioTTS remain warning compatibility imports. OpenAI's underlying Speech models shut down on 2027-01-06; Realtime requires a different adapter. Play.ht and Hyperbolic's old audio endpoint are retired: their legacy types remain importable and fail locally before network access.

Metadata

Release files for swarmauri_standard 0.11.0.dev13

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for swarmauri_standard 0.11.0.dev13
File Size Uploaded Provenance Actions
swarmauri_standard-0.11.0.dev13.tar.gz 258.4 kB Without provenance Details for swarmauri_standard-0.11.0.dev13.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for swarmauri_standard 0.11.0.dev13
File Interpreter ABI Platform Actions
swarmauri_standard-0.11.0.dev13-py3-none-any.whl Python 3 none any Details for swarmauri_standard-0.11.0.dev13-py3-none-any.whl

Total release size: 742.2 kB

Release files / swarmauri_standard-0.11.0.dev13.tar.gz

Download URL swarmauri_standard-0.11.0.dev13.tar.gz
Size 258.4 kB
Tags Source
SHA-256 checksum
How to use checksums
7b7cfccbe01db2a5b3d49440fb2650ac182a7d943227e48098bd7092a2f6fa43
BLAKE2b-256 checksum
How to use checksums
07027d35b8f23f550e0534c1adccc80c08db8ea9bb82120013e6e7ce314ec999
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via uv/0.13.9 {"installer":{"name":"uv","version":"0.13.9","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}

Release files / swarmauri_standard-0.11.0.dev13-py3-none-any.whl

Download URL swarmauri_standard-0.11.0.dev13-py3-none-any.whl
Size 483.7 kB
Tags Python 3
SHA-256 checksum
How to use checksums
9ed6cf62b0b699c31786feef83fe793d4926ea64af50fe08d6b60183f5a19f8e
BLAKE2b-256 checksum
How to use checksums
055fb1aaabb4c0bbf6c80e4bdcb558f4f39c041df4de1e2003804c8f6da57e46
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via uv/0.13.9 {"installer":{"name":"uv","version":"0.13.9","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}

Security for swarmauri_standard 0.11.0.dev13

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release
Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page