Feat/add infra script #5
Merged
Conversation
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
…gent - Updated the error handling logic to provide clearer feedback when parsing errors occur during hotel search queries. - Enhanced the response formatting in the hotel search agent to ensure proper separation of actions and inputs, following the defined usage rules. - Removed outdated log files and improved the clarity of prompts for better user guidance in hotel search tasks.
- Increased the maximum iterations for tool calls from 2 to 3 to allow for recovery iterations if needed. - Added critical rules in the hotel search assistant prompt to ensure the `search_vector_database` tool is called only once per user request, improving efficiency. - Updated filtering instructions to ensure only hotels matching all user criteria are included in the final response. - Enhanced response formatting guidelines to require complete hotel details in the final answer, ensuring clarity and user satisfaction.
…h agent - Introduced a new log file to capture detailed execution information for hotel search queries, enhancing observability. - Updated the hotel search assistant prompt to enforce strict filtering rules, ensuring only hotels that match both city and specified amenities are returned. - Improved response formatting guidelines to require comprehensive hotel details in the final answer, promoting clarity and user satisfaction.
…d result handling - Increased the maximum iterations for the LlamaIndex agent from 6 to 10 to allow for more reasoning steps and reduce iteration limit failures. - Added a mechanism to clear the last result in the LandmarkSearchEvaluator to prevent indexing bugs between queries. - Removed outdated log file to streamline logging and improve performance. - Updated the landmark search assistant prompt to enforce strict formatting rules for presenting results, ensuring clarity and user satisfaction.
…queries - Added logic to clear the cached state of the agent's last result before executing new queries, preventing potential indexing bugs. - Ensured that each query starts with a clean slate to enhance the reliability of the agent's responses.
… response clarity - Updated the input cleaning process to prioritize longer artifact patterns, ensuring more effective removal of unwanted text. - Enhanced the handling of empty input cases to include additional indicators, improving the agent's robustness. - Improved the error handling logic to provide more informative responses when parsing errors occur, aiding user understanding. - Revised the agent prompt to enforce a clear structure for responses, ensuring that final answers are provided after observations. - Added a new log file to capture detailed execution information for flight search queries, enhancing observability and debugging capabilities.
… dynamic data handling - Modified .gitignore to ignore all .mcp.json files. - Updated flight search agent queries to replace static booking IDs and departure dates with dynamically generated placeholders, improving flexibility and clarity in responses. - Added notes to clarify that certain data points are dynamically calculated based on the current date.
Resolve conflict by converting couchbase-infrastructure to submodule. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
- Update couchbase-infrastructure submodule with recreate_if_exists feature - Set recreate_if_exists=True in setup_infra.py to get fresh passwords - Add Google Colab notebook with infrastructure setup automation - Fix connection string protocol detection (auto-add couchbases://) - Fix certificate file validation (support .pem, .crt, .cer, .txt) - Fix .env file generation with proper variable validation - Add debugging output for environment verification Fixes: - ERROR: 'conn_string' - missing protocol prefix - Authentication failure - password retrieval issue - Certificate path corruption 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
Updated submodule to include analytics permissions for database users. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
- Updated setup script to use create_developer_pro_cluster - Synced couchbase-infrastructure submodule with Developer Pro support - Developer Pro cluster enables Analytics service for agentc init - Provides full database permissions including analytics management 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
- Deleted old flight_search_agent_tutorial.ipynb - Renamed flight_search_colab files to flight_search_agent_tutorial - Updated to use create_developer_pro_cluster with Analytics support 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
Updated to include __init__.py fix for cluster function exports. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
Copied flight_search_agent_tutorial.json to .ipynb to sync changes. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
Updated submodule to remove unsupported analytics privileges. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
Updated submodule to grant cluster-wide bucket access. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
Added wget command to download .agentcignore file which is needed for agentc to properly ignore files during catalog operations. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
Changed flight_search_agent tools to use lazy connection pattern with _get_cluster() that only connects when tools are called, not at import time. This prevents timeout errors when agentc indexes tools without valid credentials. Files updated: - notebooks/flight_search_agent_langraph/tools/lookup_flight_info.py - notebooks/flight_search_agent_langraph/tools/search_airline_reviews.py - notebooks/flight_search_agent_langraph/tools/retrieve_flight_bookings.py - notebooks/flight_search_agent_langraph/tools/save_flight_booking.py Note: landmark_search_agent and hotel_search_agent already had lazy connection patterns using get_cluster_connection(). 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
Removed leftover checks for undefined 'cluster' variable after refactoring to lazy connection pattern. The tools now properly use _get_cluster() which raises exceptions on connection failure. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.