Skip to main content

glytos

CI PyPI License

The official Glytos server SDK for Python.

Call the Glytos API from your backend with an API key. Build agents once and run them as text or as voice: hold a threaded conversation, stream a reply as it is written, place phone calls, mint browser web-call tokens, manage numbers, and verify webhooks.

Never ship an API key to the browser. For in-browser voice, use the @glytos/web package with a short-lived token you mint here.

Install

pip install glytos

Quickstart

from glytos import Glytos

glytos = Glytos(api_key="gly_...")

# List your agents
agents = glytos.agents.list()

# Mint a web-call token for the browser
token = glytos.calls.web_token(workflow_uuid=agents[0]["uuid"])
print(token["token"], token["ws_url"])

Use it as a context manager to close the HTTP connection cleanly:

with Glytos(api_key="gly_...") as glytos:
    overview = glytos.request("GET", "/analytics/overview")

Text conversations

An agent is one definition; nothing forces it to do both text and voice. For text, a thread holds the conversation and a run is one turn on it:

thread = glytos.threads.create(agent=agent_uuid)
run = glytos.threads.runs.create(thread, "What are your opening hours?")
print(run["messages"][-1]["content"])

Stream a long answer instead of waiting for it:

for event in glytos.threads.runs.stream(thread, "Summarise the policy"):
    if event.type == "token":
        print(event.delta, end="", flush=True)
    elif event.type == "done":
        print()

Extra context for one turn only, applied below the agent's own instructions and never saved to it:

glytos.threads.runs.create(
    thread,
    "Rate this transcript",
    instructions="Score 1-5 and reply as JSON.",
)

Everything above has an async twin on AsyncGlytos (async for over the stream).

Resources

Namespace Methods
glytos.agents (alias workflows) list, retrieve, create, rename, publish, promote, duplicate, archive, delete, templates, export, move_to_folder, remove_from_folder, versions, start_session, send_message, stream_message, run_text
glytos.threads create, retrieve, messages.create, messages.list, runs.create, runs.stream
glytos.folders list, create, rename, delete
glytos.imports sources, create, assistant
glytos.chat token, messages, stream, upload_file
glytos.calls create, list, retrieve, web_token, control
glytos.phone_numbers search, list, providers, provision, import_number, instant, assign, release
glytos.knowledge_base list_documents, create_document, upload_document, search
glytos.vector_stores list, create, retrieve, delete, upload_document
glytos.tools list, create, update, delete
glytos.campaigns list, create, retrieve, start, sync_contacts
glytos.sessions list
glytos.analytics overview
glytos.webhooks list, create, update, delete, events, deliveries, redeliver, verify

agents and workflows are the same resource under two names: the product calls them agents, the API path is /workflows. Either works.

Text and voice are separate

An agent is one definition. Nothing forces it to do both:

  • A text agent needs only threads (or chat for a browser widget).
  • A voice agent adds calls, phone_numbers and campaigns.
  • The same agent can do both, if you want it to.

Any endpoint without a dedicated helper is one call away with glytos.request(method, path, json=..., params=...), or glytos.stream(method, path, json=...) for a Server-Sent Events one.

Errors

Non-2xx responses raise a GlytosError with the API error code, HTTP status, and the request_id:

from glytos import GlytosError

try:
    glytos.workflows.retrieve("missing")
except GlytosError as err:
    print(err.status, err.code, err.message)

Webhooks

Verify a delivery came from Glytos before trusting it. Pass the raw request body, the X-Glytos-Signature header, and your endpoint secret:

from glytos import verify_webhook

# e.g. in a Flask/FastAPI handler
ok = verify_webhook(raw_body, request.headers["X-Glytos-Signature"], webhook_secret)
if not ok:
    abort(400)

License

MIT

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

glytos-0.2.1.tar.gz (15.2 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

glytos-0.2.1-py3-none-any.whl (15.4 kB view details)

Uploaded Python 3

File details

Details for the file glytos-0.2.1.tar.gz.

File metadata

  • Download URL: glytos-0.2.1.tar.gz
  • Upload date:
  • Size: 15.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.14

File hashes

Hashes for glytos-0.2.1.tar.gz
Algorithm Hash digest
SHA256 05219901ff001b5a8b3cd36e9d0d6257a9efaac51fcf002be35754e8f4abc3fa
MD5 9fbdabb7f401a8edad3ee5b7d9367428
BLAKE2b-256 c18d00f873568a801aa819a1206cff8458883cf492f11cee11eeadf7153158fc

See more details on using hashes here.

Provenance

The following attestation bundles were made for glytos-0.2.1.tar.gz:

Publisher: release.yml on Glytos/glytos-sdk-python

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file glytos-0.2.1-py3-none-any.whl.

File metadata

  • Download URL: glytos-0.2.1-py3-none-any.whl
  • Upload date:
  • Size: 15.4 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.14

File hashes

Hashes for glytos-0.2.1-py3-none-any.whl
Algorithm Hash digest
SHA256 20a6eee7a6481e266d81bc7a17ecc7e7e37e895e6314b4c0a4fbaa8f42f39a3d
MD5 2d97c9c7e07a29830608043fb0c8ba98
BLAKE2b-256 c44d03888496e6301fab230091e517a6f9791452b9cac2a1f2811d52e03f6a51

See more details on using hashes here.

Provenance

The following attestation bundles were made for glytos-0.2.1-py3-none-any.whl:

Publisher: release.yml on Glytos/glytos-sdk-python

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Release history Release notifications | RSS feed

0.5.0

2 files

0.4.0

2 files

0.3.0

2 files

This release

0.2.1 This release

2 files

0.2.0

2 files

0.1.1

2 files

0.1.0

2 files

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