Merged
Conversation
Introduces a Redis pub/sub signal mechanism to trigger code indexing after bundle uploads, with a listener in watch_index.py that spawns ingest jobs. The VSCode extension now supports configuration for a dedicated auth backend URL and shared token, enabling non-interactive authentication flows for team deployments. Also bumps the extension version to 0.1.49.
Removed the ID column and hid advanced hash and graph columns in the admin collections table for a cleaner interface. Updated table and cell styling for improved readability. Added a new 'Team' section to the VSCode extension settings for shared authentication, and bumped the extension version to 0.1.50.
Added .cursorrules and GEMINI.md containing detailed rules and best practices for AI agents using Context-Engine MCP tools. Updated .gitignore to stop ignoring .cursorrules and GEMINI.md.
Code Review SummaryStatus: No Issues Found | Recommendation: Merge Files Reviewed (9 files)
|
🤖 Augment PR SummarySummary: This PR reduces redundant/incorrect indexing work and improves remote/team auth ergonomics. Changes:
Technical Notes: The watcher’s Redis listener spawns ingest in a background subprocess and uses a lock to skip overlapping signals; auth URL normalization now preserves path-prefixed backends (e.g. 🤖 Was this summary useful? React with 👍 or 👎 |
Expanded documentation in .cursorrules, GEMINI.md, and SKILL.md to emphasize that 'symbol_graph' is the default and always-available tool for code graph queries, and clarified when to use or avoid 'neo4j_graph_query', grep, and file reading. Updated mcp version in pyproject.toml and requirements.txt for consistency. Added uv.lock for dependency management.
Updated SKILL.md and CLAUDE.example.md to explicitly discourage using grep, Read File, and similar filesystem tools for code exploration. Added clear guidance to always prefer MCP tools (repo_search, symbol_graph, etc.) for exploration, debugging, and structural queries, reserving literal search/file reads only for exact string matches or files already located via MCP. Clarified tool availability and fallback behavior for symbol_graph and neo4j_graph_query.
Python scripts now detect git repositories by presence of .git directories, preventing subdirectories from being misclassified as repos. The VSCode extension refactors backend URL normalization into a shared function, reducing duplication and improving consistency in auth-related flows.
Collaborator
Author
|
augment review |
![augmentcode[bot]](https://avatars.githubusercontent.com/in/1027498?s=60&v=4){kind=small}
| const configuredAuthBackendUrl = (settings.get('authBackendUrl') || '').trim(); | ||
| const configuredAuthToken = (settings.get('authSharedToken') || '').trim(); | ||
|
|
||
| let backendUrl = normalizeBackendUrl(explicitBackendUrl || configuredAuthBackendUrl || endpoint); |
There was a problem hiding this comment.
runAuthLoginFlow() prefers explicitBackendUrl over the new contextEngineUploader.authBackendUrl, so if callers pass the regular endpoint as the explicit URL this setting may never take effect. Can you confirm the intended precedence here?
🤖 Was this useful? React with 👍 or 👎
Enhanced multi-repo mode to prefer directories with .git, but fall back to treating all subdirectories as repos if none are found, improving resilience in K8s/container environments. Updated documentation and logging to clarify behavior. Also canonicalized working directory in CLI for macOS compatibility and fixed file permissions for the VSCode extension publish script.
Introduces a new 'Status' section in the settings webview to display live indexing progress and system status. Adds UI components, styles, and polling logic to fetch and display real-time status from the backend. The default section is now 'Status', and the webview's CSP is updated to allow connections to the configured endpoint.
![kilo-code-bot[bot]](https://avatars.githubusercontent.com/in/2193792?s=60&v=4){kind=small}
In qdrant.py, remote embedding now chunks large batches to respect the service's max batch size, improving reliability for large inputs. watch_index.py adds a check to skip signals with empty workspace_path, preventing unnecessary processing. In auth_utils.js, backend URL selection for login and logout now consistently prefers the configured authBackendUrl, ensuring settings override and alignment between login and logout flows.
Split the reset logic into modular helper functions for full and partial resets, improving readability and maintainability. The new structure separates concerns for stopping services, building containers, clearing caches, and starting the indexer, and provides clearer progress output. Also, minor cleanup in docker-compose.yml.
m1rl0k
added a commit
that referenced
this pull request
Mar 1, 2026
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.
No description provided.