[Fix] Tasks stall in repeated memory preflight calls - #2032
Draft
roomote-roomote[bot] wants to merge 1 commit into
Draft
[Fix] Tasks stall in repeated memory preflight calls#2032roomote-roomote[bot] wants to merge 1 commit into
roomote-roomote[bot] wants to merge 1 commit into
Conversation
Contributor
|
1 issue outstanding. See task
Reviewed 917d03c |
![roomote-community[bot]](https://avatars.githubusercontent.com/u/211522643?s=60&v=4){kind=small}
| const initialRecallGuidance = | ||
| serverId === BRAIN_MCP_ID | ||
| ? `The Brain-specific guidance below owns the required initial recall. Do not perform an additional generic memory preflight.` | ||
| : `At the start of each substantive task, make exactly one normal ${displayName} tool call before any other context or work tool call. Use the server's most appropriate read, recall, or search tool to retrieve relevant preferences, prior decisions, conventions, and lessons, then wait for the result before continuing. Once that call returns, the initial memory preflight is satisfied for the task; tool results, runtime continuations, retries, and continued work on the same topic do not reset it. Skip it only for greetings, simple calculations or transformations, exact actions requiring no contextual judgment, or follow-ups already covered by memory recall in the current conversation.`; |
Contributor
There was a problem hiding this comment.
The replacement prompt text removes phrases that current consumer tests still assert, but this PR updates only the direct @roomote/types tests. As a result, the focused @roomote/cloud-agents tests fail in fast-agent-integration-broker.test.ts and fast-agent-prompt.test.ts, and the Worker bootstrap test also fails. Update those expectations to the new preflight contract so the affected packages pass CI.
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.
What changed
queryas satisfying preflight across tool results, runtime continuations, retries, and continued work on the same topic.Why this change was made
Generic memory guidance allowed any read tool while the appended Brain guidance separately required
query. Those two standing preflight owners could be re-evaluated after every tool result, causing an agent to alternate memory tools indefinitely instead of beginning the requested work.Impact
Tasks and conversations retain their initial memory recall, but completing that recall now advances the agent into actual work rather than reopening the same preflight on each continuation. Legitimate targeted memory follow-ups remain available.