Repository navigation
perf(agents): an agent does as well with the short MCP answer schema - #341
Merged
Merged
Conversation
Haiku 5.5 over memo's MCP server on the synthetic v3 store, 21 tasks twice per arm, scored exactly: 39/42 with the short listing (3,323 characters for the three tools) and 39/42 with the full one (17,278), on 2.49 M against 5.29 M input tokens. The arms differ only in the outputSchema of memo_read and memo_write, shown in the tool description, since the Messages API has no output-schema field. Spend $0.83. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Task
Check that the short MCP answer schema in cli-core costs an agent nothing before it ships.
What
performance/agents/mcp-short-listing: the eval script, its 21 tasks, the two listings and the raw runs. Haiku 5.5 on memo's MCP server over a synthetic store, two runs per arm, exact scoring.🤖 Generated with Claude Code