feat: support compose-managed multimodal subagents - #129
Merged
Conversation
![kit-code-agent[bot]](https://avatars.githubusercontent.com/in/4797102?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Optional native-image capability parsing discards valid context-window metadata from legacy/custom catalogs, disabling automatic context-threshold compaction for those text sessions. Preserve the existing context-discovery behavior independently of native-generation eligibility.
danielkov
force-pushed
the
feat/compose-files-phase-2
branch
from
September 8, 2026 17:10
2e88405 to
7f97ea1
Compare
danielkov
force-pushed
the
feat/compose-files-phase-3
branch
from
September 8, 2026 17:10
eeedaff to
d84baa2
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Adds typed image attachments and native generated-image output to compose-managed subagents, with session-authorized File values and strict output binding. Enables an end-to-end read/transform/edit/return pipeline while keeping compose as the only exposed tool.
Stacked on #127; this PR intentionally targets
feat/compose-files-phase-2.Impact
Subagent, continuation, and fork calls can receive explicit managed-image attachments. Returned images remain available after child close, and native/fallback forks inherit file authority without granting access to unrelated sessions. Existing text-only output contracts retain their fallback behavior.
Technical details
x-kit-image-indexenables explicit selection from multiple outputs after every occurrence passes validation and budgets.Implements Phase 3 of the compose-managed-files plan. Shared TUI/Markdown media presentation remains Phase 4.