arcade-mcp/toolkits/confluence/evals/eval_page.py
Eric Gustin 6bc6f0c574
Add Confluence toolkit (#406)
12 tools for the new Confluence toolkit.

| Name | Description |

|----------------------------------|------------------------------------------------------------------------------------|
| Confluence.CreatePage | Create a new page at the root of the given
space. |
| Confluence.UpdatePageContent | Update a page's content. |
| Confluence.RenamePage | Rename a page by changing its title. |
| Confluence.GetPage | Retrieve a SINGLE page's content by its ID or
title. |
| Confluence.GetPagesById | Get the content of MULTIPLE pages by their
ID in a single efficient request. |
| Confluence.ListPages | Get the content of multiple pages by their ID.
|
| Confluence.ListAttachments | List attachments in a workspace. |
| Confluence.GetAttachmentsForPage | Get attachments for a page by its
ID or title. |
| Confluence.SearchContent | Search for content in Confluence. |
| Confluence.GetSpace | Get the details of a space by its ID or key. |
| Confluence.ListSpaces | List all spaces sorted by name in ascending
order. |
| Confluence.GetSpaceHierarchy | Retrieve the full hierarchical
structure of a Confluence space as a tree structure.|

### Confluence Clients
Confluence has deprecated most of their V1 endpoints, so most of the
tools use V2. However, we still need a V1 client to support search. The
V1 search API has not been deprecated yet, because there is no V2
equivalent. But we need to be aware in the future for when this
deprecation happens.


### Future work
* Content of pages are returned in the Confluence `storage` format. This
is the format that Confluence uses to store pages internally. We should
understand the storage format more deeply, and write utility function to
transform this format into plain text.
* Better protections against extremely large pages. I've tested up to
6,000 word pages.
* Tools for blog posts
* Tool for getting the children of a page. (`get_space_hierarchy` will
suffice for now)
* Allow for numerical titles
2025-05-22 13:18:19 -07:00

207 lines
6.4 KiB
Python

from arcade.sdk import ToolCatalog
from arcade.sdk.eval import (
EvalRubric,
EvalSuite,
ExpectedToolCall,
tool_eval,
)
from arcade.sdk.eval.critic import BinaryCritic, SimilarityCritic
import arcade_confluence
from arcade_confluence.enums import PageUpdateMode
from arcade_confluence.tools import (
create_page,
get_page,
get_pages_by_id,
rename_page,
update_page_content,
)
from evals.conversations import create_page_1, get_space_hierarchy_1
from evals.critics import ListCritic
# Evaluation rubric
rubric = EvalRubric(
fail_threshold=0.85,
warn_threshold=0.95,
)
catalog = ToolCatalog()
catalog.add_module(arcade_confluence)
@tool_eval()
def confluence_get_page_eval_suite() -> EvalSuite:
suite = EvalSuite(
name="Confluence get_page tool evaluation",
system_message="You are an AI assistant with access to Confluence tools.",
catalog=catalog,
rubric=rubric,
)
suite.add_case(
name="Get page by ID",
user_message="Get page 65816",
expected_tool_calls=[ExpectedToolCall(func=get_page, args={"page_identifier": 65816})],
rubric=rubric,
critics=[BinaryCritic(critic_field="page_identifier", weight=1)],
)
suite.add_case(
name="Get page by title",
user_message="Get my 'Poem - May 24th' page",
expected_tool_calls=[
ExpectedToolCall(func=get_page, args={"page_identifier": "Poem - May 24th"})
],
rubric=rubric,
critics=[
BinaryCritic(critic_field="page_identifier", weight=1),
],
)
suite.add_case(
name="Get page based on previous conversation",
user_message=(
"Get the content of my daily note. You MUST use the page's title when getting the page."
),
expected_tool_calls=[ExpectedToolCall(func=get_page, args={"page_identifier": "10-03-98"})],
rubric=rubric,
critics=[
BinaryCritic(critic_field="page_identifier", weight=1),
],
additional_messages=get_space_hierarchy_1,
)
return suite
@tool_eval()
def confluence_get_multiple_pages_by_id_eval_suite() -> EvalSuite:
suite = EvalSuite(
name="Confluence get_multiple_pages_by_id tool evaluation",
system_message="You are an AI assistant with access to Confluence tools.",
catalog=catalog,
rubric=rubric,
)
suite.add_case(
name="Get multiple pages by ID",
user_message="Get 98418, 4685837, 5242883, 5275653, 5242903, 5242913 pages",
expected_tool_calls=[
ExpectedToolCall(
func=get_pages_by_id,
args={"page_ids": ["98418", "4685837", "5242883", "5275653", "5242903", "5242913"]},
)
],
rubric=rubric,
critics=[
ListCritic(critic_field="page_ids", order_matters=False, weight=1),
],
)
suite.add_case(
name="Get multiple pages by ID with existing conversation",
user_message=("Get the content of all pages in the space except 4685837"),
expected_tool_calls=[
ExpectedToolCall(
func=get_pages_by_id,
args={"page_ids": ["98418", "5242883", "5275653", "5242903", "5242913"]},
)
],
rubric=rubric,
critics=[
ListCritic(critic_field="page_ids", order_matters=False, weight=1),
],
additional_messages=get_space_hierarchy_1,
)
return suite
@tool_eval()
def confluence_create_page_eval_suite() -> EvalSuite:
suite = EvalSuite(
name="Confluence create_page tool evaluation",
system_message="You are an AI assistant with access to Confluence tools.",
catalog=catalog,
rubric=rubric,
)
suite.add_case(
name="Create a page",
user_message=(
"Make a page within the softwareeng space called 'TODOs' under the 4830960 folder. "
"Within it, make a list of all the TODOs for the day which are "
"1. Write a blog post about the future of AI"
"2. Write an agent that calls my mom at 5:30 every Friday evening"
),
expected_tool_calls=[
ExpectedToolCall(
func=create_page,
args={
"space_identifier": "softwareeng",
"title": "TODOs",
"content": "1. Write a blog post about the future of AI\n2. Write an agent that calls my mom at 5:30 every Friday evening", # noqa: E501
"parent_id": "4830960",
},
)
],
rubric=rubric,
critics=[
BinaryCritic(critic_field="space_identifier", weight=1 / 4),
BinaryCritic(critic_field="title", weight=1 / 4),
SimilarityCritic(critic_field="content", weight=1 / 4),
BinaryCritic(critic_field="parent_id", weight=1 / 4),
],
)
return suite
@tool_eval()
def confluence_update_page_eval_suite() -> EvalSuite:
suite = EvalSuite(
name="Confluence update page tools evaluation",
system_message="You are an AI assistant with access to Confluence tools.",
catalog=catalog,
rubric=rubric,
)
suite.add_case(
name="Update page content",
user_message="Thanks, now append '3. Walk the dog' to the end of the list",
expected_tool_calls=[
ExpectedToolCall(
func=update_page_content,
args={
"page_identifier": "5439489",
"content": "3. Walk the dog",
"update_mode": PageUpdateMode.APPEND,
},
)
],
rubric=rubric,
critics=[
BinaryCritic(critic_field="page_identifier", weight=1 / 3),
SimilarityCritic(critic_field="content", weight=1 / 3),
BinaryCritic(critic_field="update_mode", weight=1 / 3),
],
additional_messages=create_page_1,
)
suite.extend_case(
name="Rename page",
user_message="Actually, rename it to 'My TODOs'",
expected_tool_calls=[
ExpectedToolCall(
func=rename_page,
args={"page_identifier": "5439489", "title": "My TODOs"},
),
],
rubric=rubric,
critics=[
BinaryCritic(critic_field="page_identifier", weight=1 / 2),
BinaryCritic(critic_field="title", weight=1 / 2),
],
)
return suite