codex/codex-rs/ext
pakrym-oai ce0323f2f2 Support typed content in contextual prompt fragments (#54776)
## Why

Contextual fragments and extension-owned world-state sections currently return strings, limiting them to text content.

## What changed

- Return `ContentItem` from fragment bodies and rendering, preserving classification and attribution. Apply text markers only to `InputText`; pass other content through unchanged.
- Allow extension prompt and world-state contributions to carry typed content, and preserve media in context-window hints.
- Hash non-text world-state fragments by their serialized content and role while retaining the existing text hashing behavior.
- Add shared content-to-tool-output conversion that preserves image references, image detail, and audio.

## Testing

Add coverage for typed world-state hashing and preservation of image content and attribution. Expand code-mode output tests to cover input text, output text, file-backed and inline images, and audio.

GitOrigin-RevId: 2ed00d466a3f169b151202f3eef71e667ea64be1
History 2026-10-17 04:02:21 +00:00
..
agent Expose and persist turn lineage across the app server (#51415) 2026-10-06 16:56:39 +00:00
agent-message-board Store generated image items as attachment references (#54398) 2026-10-15 19:51:22 +00:00
connectors Rename plugin MCP and app extension APIs (#46564) 2026-09-19 01:45:51 +00:00
extension-api Support typed content in contextual prompt fragments (#54776) 2026-10-17 04:02:21 +00:00
git-attribution Support typed content in contextual prompt fragments (#54776) 2026-10-17 04:02:21 +00:00
goal Store generated image items as attachment references (#54398) 2026-10-15 19:51:22 +00:00
guardian-reviewer Share Guardian image deduplication and checkpoint restoration (#54278) 2026-10-15 12:42:15 +00:00
guardian-v2 Support typed content in contextual prompt fragments (#54776) 2026-10-17 04:02:21 +00:00
history-notes Store generated image items as attachment references (#54398) 2026-10-15 19:51:22 +00:00
image-generation Record local image inputs on completed image-generation items (#54420) 2026-10-15 21:30:56 +00:00
items Record local image inputs on completed image-generation items (#54420) 2026-10-15 21:30:56 +00:00
mcp Acquire and cache opt-in executor capability inventories via V2 (#53736) 2026-10-13 18:15:50 +00:00
memories Support typed content in contextual prompt fragments (#54776) 2026-10-17 04:02:21 +00:00
multi-env-copy-file Add an opt-in tool for copying files between environments (#54674) 2026-10-16 18:39:31 +00:00
queue Expose and persist turn lineage across the app server (#51415) 2026-10-06 16:56:39 +00:00
skills Support typed content in contextual prompt fragments (#54776) 2026-10-17 04:02:21 +00:00
web-search Add opt-in output token replay for OpenAI requests (#52742) 2026-10-10 01:25:00 +00:00