219 files, 2.0 MB, untracked in souveraine/docs and existing nowhere else. The volume is at 100% with no snapshots.
1.3 KiB
1.3 KiB
| description | priority | phase | status | depends_on |
|---|---|---|---|---|
| Standardized formatting for pasted content to distinguish it from agent output and subconscious surfacing | medium | scoping | open |
Pasted Content Formatting
Goal
When the user pastes context (conversation excerpts, code blocks, reference material), the agent should recognize it as pasted content and format it consistently — distinguishable from the user's own words, the agent's output, and subconscious surfacing.
Current Problem
Pasted content appears inline with no visual distinction. Large pastes blend into the conversation, making it hard to tell what the user wrote vs what they pasted for context.
Agreed Design Elements (from session 2026-05-18)
- Pasted content should have a visual marker or container
- Formatting conventions need to be consistent across all agent interactions
- The agent should be able to detect when content is being pasted vs composed
Open Questions
- What visual container? Blockquote style? Code fence with a label? Collapsible section?
- How does the agent detect pasted vs composed text? Heuristic (length/formatting) or explicit marker?
- Should the detection be handled in the TUI or in the agent's prompt processing?
Related
docs/INDEX.mdfor overall doc structuredocs/bugs.mdfor related display issues