agentwatch
Lightweight OpenAI usage tracking for AgentWatch. Token counts, model, and latency are sent to AgentWatch after each response — prompts and completions never leave your server.
Install
pip install agentwatch-py
Usage
from openai import OpenAI
from agentwatch import wrap
client = wrap(
OpenAI(),
api_key="aw_live_...", # Your AgentWatch API key
agent_id="...", # Agent ID from your AgentWatch dashboard
)
# Use exactly as before — tracking happens automatically
response = client.chat.completions.create(
model="gpt-4o",
messages=[{"role": "user", "content": "Hello"}],
)
# Streaming works too
for chunk in client.chat.completions.create(
model="gpt-4o",
messages=[{"role": "user", "content": "Hello"}],
stream=True,
):
pass # process chunk as normal
What gets sent to AgentWatch
Only usage metadata — never your prompts or completions:
{
"agent_id": "...",
"model": "gpt-4o",
"prompt_tokens": 12,
"completion_tokens": 34,
"latency_ms": 843,
"success": true
}
Options
| Argument | Type | Description |
|---|---|---|
api_key |
str |
Your AgentWatch API key (aw_live_...) |
agent_id |
str |
Agent ID from your dashboard |
ingest_url |
str |
Override the ingest URL (optional) |
Find your API key and agent IDs at getagentwatch.com/dashboard/integrations.
Metadata
Release files for agentwatch-py 0.2.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| agentwatch_py-0.2.0.tar.gz | 4.0 kB | Without provenance | Details for agentwatch_py-0.2.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| agentwatch_py-0.2.0-py3-none-any.whl | Python 3 | none | any | Details for agentwatch_py-0.2.0-py3-none-any.whl |
Total release size: 8.3 kB
Release files / agentwatch_py-0.2.0.tar.gz
| Download URL | agentwatch_py-0.2.0.tar.gz |
|---|---|
| Size | 4.0 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
7876d3594c83d1ad452ddb31b5404f17c7eaf44f0e0cf07b4bd37ed8ee336de2
|
|
BLAKE2b-256 checksum How to use checksums |
d323bfd0c6bfb7f45462a9ae3b66a64b4fe5cb5603485104b2679cee074f88cd
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.3
|
Release files / agentwatch_py-0.2.0-py3-none-any.whl
| Download URL | agentwatch_py-0.2.0-py3-none-any.whl |
|---|---|
| Size | 4.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
89f4762eb027e2cca45e2ee9a18c3e69098d4609835501835cc96bee704ce824
|
|
BLAKE2b-256 checksum How to use checksums |
85242b199f3eb856959391165f56c842ff7157772eb6cf69dff2e01e32ba8eef
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.3
|
Security for agentwatch-py 0.2.0
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.