chore(DATAGO-124643): Order agent card display names in alphabetical order

[Hugo-Pare]

What is the purpose of this change?

Implements alphabetical sorting for agent cards returned by the /api/v1/agentCards endpoint. This ensures that agent dropdowns and lists in the frontend display agents in a consistent, user-friendly alphabetical order based on their display names.

How was this change implemented?

Backend Changes:

• Modified src/solace_agent_mesh/gateway/http_sse/routers/agent_cards.py
  • Added DISPLAY_NAME_EXTENSION_URI constant for the display name extension
  • Created _get_agent_display_name() helper function to extract display names from agent card extensions, falling back to agent name if not present
  • Added case-insensitive alphabetical sorting before returning agent cards (lines 204-207)

Test Assets:

• Created examples/agents/test_sorting_agents.yaml with 12 test agents featuring diverse display names (numbers, emojis, special characters, mixed case) for manual testing
• Created tests/integration/apis/test_agent_cards_sorting.py integration test covering 20 edge cases including:
  • Leading spaces, numbers, special characters (@, &, _, -, .)
  • Mixed case and all-caps names
  • International characters (accented letters, Spanish ñ, Chinese characters)
  • Emojis and Unicode symbols

Key Design Decisions

• Sorting at API level: Implemented sorting in the backend API endpoint rather than frontend to ensure consistency across all consumers of the API
• Case-insensitive sorting: Uses Python's .lower() for case-insensitive comparison, so "Apple Agent" and "ALPHA AGENT" sort logically together
• Display name extraction: Properly handles the A2A protocol's display name extension while gracefully falling back to internal agent names

How was this change tested?

• Unit tests: Created comprehensive integration test with 20 edge cases covering numbers, special characters, international characters, emojis, and various Unicode edge cases
• Integration tests: test_agent_cards_endpoint_returns_alphabetically_sorted_agents tests the full flow: creates test agents → registers in agent registry → calls API endpoint → verifies alphabetical order
• Manual testing: Created examples/agents/test_sorting_agents.yaml with 12 sample agents for visual verification in the frontend
• Known limitations: Sorting behavior for emojis and non-Latin scripts follows Python's default string sorting (Unicode code point order)

Is there anything the reviewers should focus on/be aware of?

• The sort order for special characters follows Python's default string comparison (e.g., underscore _ sorts before letters, emojis sort after ASCII)
• The integration test sets up its own AgentRegistry instance rather than using a fixture, as the WebUIBackendFactory doesn't pre-configure an agent registry - reviewers should verify this approach aligns with testing patterns
• Display names are extracted from the https://solace.com/a2a/extensions/display-name extension; agents without this extension fall back to using their internal name field

References

Default agent in Projects

Agent selection in Chat

[sonarqube-solacecloud]

Quality Gate passed

Issues
0 New issues
0 Fixed issues
0 Accepted issues

Measures
0 Security Hotspots
90.9% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube