- Add GitStore class wrapping dulwich for memory file versioning
- Auto-commit memory changes during Dream consolidation
- Add /dream-log and /dream-restore commands for history browsing
- Pass tracked_files as constructor param, generate .gitignore dynamically
Add "Solutions" category to consolidate prompt so trial-and-error
workflows that reach a working approach are captured in history for
Dream to persist. Remove overly broad "debug steps" skip rule that
discarded these valuable findings.
Replace single-stage MemoryConsolidator with a two-stage architecture:
- Consolidator: lightweight token-budget triggered summarization,
appends to HISTORY.md with cursor-based tracking
- Dream: cron-scheduled two-phase processor that analyzes HISTORY.md
and updates SOUL.md, USER.md, MEMORY.md via AgentRunner with
edit_file tools for surgical, fault-tolerant updates
New files: MemoryStore (pure file I/O), Dream class, DreamConfig,
/dream and /dream-log commands. 89 tests covering all components.
Different channels could theoretically share the same chat_id.
Check both channel and chat_id to avoid cross-channel reply issues.
Co-authored-by: layla <111667698+04cb@users.noreply.github.com>
When the message tool is used to send a message to a different chat_id
than the current conversation, it was incorrectly including the default
message_id from the original context. This caused channels like Feishu
to send the message as a reply to the original chat instead of creating
a new message in the target chat.
Changes:
- Only use default message_id when chat_id matches the default context
- When targeting a different chat, set message_id to None to avoid
unintended reply behavior
Address PR review feedback by avoiding an async method reference as the OpenAI client api_key.
Initialize the client with a placeholder key, refresh the Copilot token before each chat/chat_stream call, and update the runtime client api_key before dispatch.
Add a regression test that verifies the client api_key is refreshed to a real string before chat requests.
Generated with GitHub Copilot, GPT-5.4.
Implement the real GitHub device flow and Copilot token exchange for the GitHub Copilot provider.
Also route github-copilot models through a dedicated backend and strip the provider prefix before API requests.
Add focused regression coverage for provider wiring and model normalization.
Generated with GitHub Copilot, GPT-5.4.
Remove the fragile barrier-based synchronization that could cause
deadlock when the second request is scheduled first. Instead, rely
on the session lock for serialization and handle either execution
order in assertions.
1. Fix full_url path for non-image media to require AES key and skip download when missing,
instead of persisting encrypted bytes as valid media.
2. Restrict quoted media fallback trigger to only when no top-level media item exists,
not when top-level media download/decryption fails.
