llm-browser-testkit 0.2.5

LLM-driven browser test framework — define browser test scenarios in TOML with natural language steps and LLM-powered assertions
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39

# Agent-to-agent scenario — uses an A2A agent for assertions

[config.endpoints.audit_agent]
type = "a2a"
url = "http://localhost:9090"
pricing = { per_call = 0.01 }

[config.endpoints.default]
type = "llm"
url = "http://localhost:8080"
model = "deepseek"
pricing = { input_per_1m_tokens = 0.15, output_per_1m_tokens = 0.60 }

[[definitions]]
name = "no_errors"
preset = "no_error_on_page"

[[definitions]]
name = "agent_verify"
agent = "audit_agent"
task_template = "Check if the page at {url} with title '{title}' shows the following: {expected_text}"

[[test]]
name = "agent smoke test"
start_url = "/"
budget = { max_cost = 1.0, enforcement = "hard" }

[[test.steps]]
kind = "navigate"
url = "/"

[[test.steps]]
kind = "assert"
definition = "no_errors"

[[test.steps]]
kind = "agent"
agent = "audit_agent"
task = "Verify the homepage loads without errors"