Repository navigation
Let the Hub default decide how many assets a recall returns - #39
Merged
Merged
Conversation
Strategy recall sent `limit: 5` to /asset/fetch. It now sends only the prompt text, so the Hub's default applies (two assets today). Checked against the live Proxy: three sample prompts each came back with two assets in 2.3-3.7s, and each still had one priming would inject. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Merged
forrestlinfeng
added a commit
that referenced
this pull request
Oct 10, 2026
Strategy recall lets the Hub default decide how many assets come back instead of asking for five (#39). Co-authored-by: forrestlinfeng <linfengchen@tencent.com> 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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Strategy recall no longer sends
limit: 5to/asset/fetch; the request carries only the prompt text, so the Hub's default count applies (two assets today).Trade-off
With two candidates instead of five, a prompt whose first two candidates are both filtered out (too few steps, over 4000 chars, low similarity, or already injected this session) now injects nothing, where the 3rd–5th might have fit.
/evolver-statusshows this asall_filteredwith the reasons, so it is visible if it happens.Test plan
npm test: 101/101; tests now assert the request body carries nolimitlimit: 3 prompts each returned 2 assets in 2.3–3.7s, and each still had an injectable candidate🤖 Generated with Claude Code