Draft
Conversation
- Add @mention dropdown UI in input box (keyboard navigable)
- Parse {{@agentid}} tags in message content for mention badge rendering
- Backend: route @agent messages directly to persistent agents, skip workforce
- Backend: reuse persistent agents across turns (preserve toolkit state)
- Frontend: persist mention target across turns, render in input and chat bubbles
- Fix keyboard ArrowUp/Down selection in mention dropdown
Persistent agents reused across multi-turn @mention conversations were losing tool call history due to prune_tool_calls_from_memory, causing the LLM to repeat operations (e.g. get_page_snapshot) it already performed.
Move prune_tool_calls_from_memory=False from chat_service persistent agent override to browser factory directly, so all browser agents (workforce and @mention) retain tool call history.
Resolve conflicts in: - server/uv.lock (revision number, greenlet wheel format) - src/store/chatStore.ts (merge target/displayContent and executionId/projectId params) - src/components/ChatBox/BottomBox/InputBox.tsx (merge mention and trigger/expanded features) - src/components/ChatBox/index.tsx (merge mention/direct-agent features with new unified layout)
Resolve conflicts keeping feature branch mention/direct-agent additions merged with main's trigger, execution, and layout changes.
- Add ExtensionProxyWrapper support for browser_plug Chrome extension - Extension proxy managed via dedicated Extension settings page - Backend starts WebSocket server on connect, polls for extension status - Multi-tab parallelism: each sub-agent gets its own browser tab - Extension proxy mode is exclusive (no CDP fallback when connected) - Add extension_proxy_service singleton and controller endpoints - Use local camel package instead of PyPI version
Enable real-time token streaming from ChatAgent to Chrome extension UI by setting stream=True in model config and async-iterating over response chunks, sending each text delta immediately via STREAM_TEXT.
Includes sidepanel UI, background service worker with WebSocket connection to eigent backend, CDP debugger management, and auto-reconnect/tab-lock/settings-persistence robustness features.
Collaborator
Author
|
TODO：gif record for actions |
Resolve conflicts in pyproject.toml, uv.lock, UserQueryGroup.tsx, and ChatBox/index.tsx - keep branch's privacy/mention features and main's package versions.
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
Comment on lines
+56
to
+57
| f"Extension chat model configured: " | ||
| f"{config.get('model_platform')}/{config.get('model_type')}" |
Check failure
Code scanning / CodeQL
Clear-text logging of sensitive information High
Show autofix suggestion
Hide autofix suggestion
Copilot Autofix
AI 8 days ago
In general, to fix clear-text logging of sensitive information you should ensure that no secret-bearing data (API keys, passwords, tokens, full config dicts that contain them, etc.) is ever passed to logging calls. Instead, log only non-sensitive metadata (e.g., model names, platform identifiers, status flags) or sanitized versions of the data (e.g., masked values).
For this specific case, the best fix without changing functionality is to:
- Keep logging that the extension chat model has been configured and which platform/type are used (these are non-sensitive).
- Make the log string construction clearly independent of the tainted
configobject in a way that the static analyzer recognizes as safe. The simplest way is to extract just the non-sensitive values into local variables and log only those, ensuring we never logconfigitself or any field that might be sensitive.
Concretely, in backend/app/service/extension_chat_service.py, in configure_model:
- Introduce local variables, e.g.
platform = config.get('model_platform')andmodel_type = config.get('model_type'). - Use these variables in the
logger.infomessage, rather than referring toconfiginline in the formatted string. - Make sure not to add any logging that includes
config,api_key, or similar sensitive values.
No other files need content changes for this specific issue, because they don’t log the config or API key.
Suggested changeset
1
backend/app/service/extension_chat_service.py
| @@ -52,9 +52,12 @@ | ||
| """ | ||
| global _model_config | ||
| _model_config = config | ||
| platform = config.get("model_platform") | ||
| model_type = config.get("model_type") | ||
| logger.info( | ||
| f"Extension chat model configured: " | ||
| f"{config.get('model_platform')}/{config.get('model_type')}" | ||
| "Extension chat model configured: %s/%s", | ||
| platform, | ||
| model_type, | ||
| ) | ||
|
|
||
|
|
Copilot is powered by AI and may make mistakes. Always verify output.
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
3 participants
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.
Related Issue
Closes #
Description
Testing Evidence (REQUIRED)
What is the purpose of this pull request?
Contribution Guidelines Acknowledgement