Skip to content

Conversation

@angaither
Copy link
Contributor

Had a few issues with the LLM between difference of making a search vs actually doing a search
It also seems like we don't need bulk order canceling.

Copy link

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull Request Overview

This PR disables additional tools in the SDK server to address LLM issues with search operations and to remove bulk order canceling functionality. The changes prevent certain Planet API functions from being exposed as AI tools where they don't work well in an AI context.

Key changes:

  • Disabled all search-related data operations (create, delete, get, list, update)
  • Disabled bulk order cancellation functionality
Comments suppressed due to low confidence (1)

src/planet_mcp/servers/sdk.py:31

  • [nitpick] The entries in _DEFAULT_IGNORE are not consistently ordered. While the set is grouped by prefix (data, orders, subscriptions, etc.), the items within each group are not alphabetically sorted. For better maintainability, consider sorting all entries alphabetically within each prefix group. For example, the data_* entries should be ordered as: data_create_search, data_delete_search, data_get_search, data_get_stats, data_list_searches, data_update_search, data_wait_asset.
_DEFAULT_IGNORE = {
    "data_wait_asset",
    "orders_wait",
    "data_create_search",
    "data_delete_search",
    "data_get_search",
    "data_get_stats",
    "data_list_searches",
    "data_update_search",
    "orders_aggregated_order_stats",
    "orders_cancel_orders",
    "subscriptions_get_results_csv",
    "subscriptions_patch_subscription",
    "subscriptions_update_subscription",
    "mosaics_get_quad_contributions",
    "destinations_patch_destination",
}

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

@asonnenschein
Copy link
Contributor

  • [nitpick] The entries in _DEFAULT_IGNORE are not consistently ordered. While the set is grouped by prefix (data, orders, subscriptions, etc.), the items within each group are not alphabetically sorted. For better maintainability, consider sorting all entries alphabetically within each prefix group. For example, the data_* entries should be ordered as: data_create_search, data_delete_search, data_get_search, data_get_stats, data_list_searches, data_update_search, data_wait_asset.
_DEFAULT_IGNORE = {
    "data_wait_asset",
    "orders_wait",
    "data_create_search",
    "data_delete_search",
    "data_get_search",
    "data_get_stats",
    "data_list_searches",
    "data_update_search",
    "orders_aggregated_order_stats",
    "orders_cancel_orders",
    "subscriptions_get_results_csv",
    "subscriptions_patch_subscription",
    "subscriptions_update_subscription",
    "mosaics_get_quad_contributions",
    "destinations_patch_destination",
}

Approving, with consideration for this nitpick comment from copilot. I agree with copilot that the _DEFAULT_IGNORE list is better implemented alphabetically.

@angaither angaither merged commit d53c994 into main Nov 18, 2025
8 checks passed
@angaither angaither deleted the ag/disable-some-tools branch November 18, 2025 16:21
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants