New Research: Supply Chain Attack on Axios Pulls Malicious Dependency from npm.Details →
Socket
Book a DemoSign in
Socket

subconscious-python

Package Overview
Dependencies
Maintainers
2
Versions
6
Alerts
File Explorer

Advanced tools

Socket logo

Install Socket

Detect and block malicious and high-risk dependencies

Install

subconscious-python - pypi Package Compare versions

Comparing version
0.1.21
to
0.1.22
+10
-2
PKG-INFO
Metadata-Version: 2.4
Name: subconscious-python
Version: 0.1.21
Version: 0.1.22
Summary: Python SDK for Subconscious AI agent framework

@@ -69,3 +69,3 @@ Author-email: Subconscious Systems <hongyin@subconscious.dev>

client = Client(
base_url="https://api.subconscious.dev",
base_url="https://api.subconscious.dev/v1", # can be omitted
api_key="your-api-key" # get it from https://subconscious.dev

@@ -77,3 +77,7 @@ )

{
"type": "function",
"name": "calculator",
"url": "https://URL_TO_CALCULATOR_TOOL/ENDPOINT", # the server url of your own tool
"method": "POST",
"timeout": 5, # seconds
"parameters": {

@@ -100,2 +104,6 @@ "type": "object",

The TIM language model will call the `calculator` tool as many times as necessary, handle excepts, compute the answer, and return the result. The agent is completed with one language model API call!
We also provide fine-grained control over the reasoning structure, tool use, and memory management. Check out the [deep research agent example](https://github.com/subconscious-systems/TIMRUN/tree/main/examples/deep_research) for more advanced usage.
## Features

@@ -102,0 +110,0 @@

+1
-1

@@ -7,3 +7,3 @@ [build-system]

name = "subconscious-python"
version = "0.1.21"
version = "0.1.22"
description = "Python SDK for Subconscious AI agent framework"

@@ -10,0 +10,0 @@ readme = "README.md"

@@ -31,3 +31,3 @@ # Subconscious Python SDK

client = Client(
base_url="https://api.subconscious.dev",
base_url="https://api.subconscious.dev/v1", # can be omitted
api_key="your-api-key" # get it from https://subconscious.dev

@@ -39,3 +39,7 @@ )

{
"type": "function",
"name": "calculator",
"url": "https://URL_TO_CALCULATOR_TOOL/ENDPOINT", # the server url of your own tool
"method": "POST",
"timeout": 5, # seconds
"parameters": {

@@ -62,2 +66,6 @@ "type": "object",

The TIM language model will call the `calculator` tool as many times as necessary, handle excepts, compute the answer, and return the result. The agent is completed with one language model API call!
We also provide fine-grained control over the reasoning structure, tool use, and memory management. Check out the [deep research agent example](https://github.com/subconscious-systems/TIMRUN/tree/main/examples/deep_research) for more advanced usage.
## Features

@@ -64,0 +72,0 @@

Metadata-Version: 2.4
Name: subconscious-python
Version: 0.1.21
Version: 0.1.22
Summary: Python SDK for Subconscious AI agent framework

@@ -69,3 +69,3 @@ Author-email: Subconscious Systems <hongyin@subconscious.dev>

client = Client(
base_url="https://api.subconscious.dev",
base_url="https://api.subconscious.dev/v1", # can be omitted
api_key="your-api-key" # get it from https://subconscious.dev

@@ -77,3 +77,7 @@ )

{
"type": "function",
"name": "calculator",
"url": "https://URL_TO_CALCULATOR_TOOL/ENDPOINT", # the server url of your own tool
"method": "POST",
"timeout": 5, # seconds
"parameters": {

@@ -100,2 +104,6 @@ "type": "object",

The TIM language model will call the `calculator` tool as many times as necessary, handle excepts, compute the answer, and return the result. The agent is completed with one language model API call!
We also provide fine-grained control over the reasoning structure, tool use, and memory management. Check out the [deep research agent example](https://github.com/subconscious-systems/TIMRUN/tree/main/examples/deep_research) for more advanced usage.
## Features

@@ -102,0 +110,0 @@

@@ -93,3 +93,3 @@ import json

def __init__(self, base_url: str = "https://api.subconscious.dev", api_key: str = None):
def __init__(self, base_url: str = "https://api.subconscious.dev/v1", api_key: str = None):
openai_client = OpenAI(

@@ -96,0 +96,0 @@ base_url=base_url, api_key=api_key

@@ -83,4 +83,4 @@ import json

try:
# if True:
# try:
if True:
# Make streaming request

@@ -127,8 +127,8 @@ response = openai_client.chat.completions.create(

except requests.exceptions.RequestException as e:
print(f"Request error: {e}")
except KeyboardInterrupt:
print("\nTest interrupted by user")
except Exception as e:
print(f"Unexpected error: {e}")
# except requests.exceptions.RequestException as e:
# print(f"Request error: {e}")
# except KeyboardInterrupt:
# print("\nTest interrupted by user")
# except Exception as e:
# print(f"Unexpected error: {e}")

@@ -135,0 +135,0 @@ def tim_streaming_legacy(model, messages, tools, reasoning_grammar_model):