Skip to content

Conversation

@zastrowm
Copy link
Member

@zastrowm zastrowm commented Nov 5, 2025

Description

Per #1067 including the args in the description is redundant as it's already included in the parameter docs which can increase the token counts. Strip args from the description string for inputSpecs

Note: This is changing the inputSpecs that we're generating which has the risk of changing agent behavior. However, I think the reduction in token counts is worth it given that argument descriptions already have this information.

Related Issues

#1067

Documentation PR

N/A - this more closely aligns our docs

Type of Change

Bug fix

Testing

How have you tested the change? Verify that the changes do not break functionality or introduce warnings in consuming repositories: agents-docs, agents-tools, agents-cli

  • I ran hatch run prepare

Checklist

  • I have read the CONTRIBUTING document
  • I have added any necessary tests that prove my fix is effective or my feature works
  • I have updated the documentation accordingly
  • I have added an appropriate example to the documentation to outline the feature, or no new docs are needed
  • My changes generate no new warnings
  • Any dependent changes have been merged and published

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

Per strands-agents#1067 including the args in the description is redundant as it's already included in the parameter docs which can increase the token counts.  Strip args from the description strings for inputSpecs
spec = tool_with_full_docstring.tool_spec
assert (
spec["description"]
== """This is the main description.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nit: So that you can present this more cleanly in code, you can use textwrap.dedent:

import textwrap
...
description = textwrap.dedent("""
    This is the main description.
    ...
""")

stripped_line = line.strip()

# Check if we're starting the Args section
if stripped_line.lower().startswith(("args:", "arguments:", "parameters:", "param:", "params:")):
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Consideration: Not saying it needs to be done for this PR, but I'm wondering if there is a reliable third party library out there that we can use to parse the docstrings. From a cursory search, I came across https://pypi.org/project/docstring-parser/. It's in preview though.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yeah - I also looked for built in ways to do this and didn't see any great ways. I think if we were starting today we'd do this differently so that we're not managing raw strings.

I'm a bit reluctant to add a library just this :(

@github-actions github-actions bot removed the size/m label Nov 5, 2025
@zastrowm zastrowm deployed to auto-approve November 5, 2025 20:35 — with GitHub Actions Active
@github-actions github-actions bot added the size/m label Nov 5, 2025
@codecov
Copy link

codecov bot commented Nov 5, 2025

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants