Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Appearance settings
Merged
Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
77 commits
Select commit Hold shift + click to select a range
d737161
Add AI agent framework infrastructure
dannon Dec 16, 2025
a3ef8f2
Fix agent imports and registrations in __init__.py
dannon Dec 16, 2025
a950d77
Add error analysis agent
dannon Dec 8, 2025
8b9cb96
Add tool recommendation agent
dannon Dec 8, 2025
90c8e9f
Add custom tool agent
dannon Dec 8, 2025
7dc41be
Add agent routing and orchestration
dannon Dec 16, 2025
3a697ea
Add configuration schema for AI agents
dannon Dec 7, 2025
78784f9
Add agent backend API and services
dannon Dec 7, 2025
9afe6d8
Add ChatGXY frontend component
dannon Dec 7, 2025
524d5e3
Add DSPy agent and test suite
dannon Dec 7, 2025
0922e5c
Improve structured output detection for vLLM/local endpoints
dannon Dec 8, 2025
654ac08
Improve type safety and exception handling in agent framework
dannon Dec 8, 2025
3d91520
Tighten exception handling in core agent framework
dannon Dec 8, 2025
862828f
Tighten exception handling in specialist agents
dannon Dec 16, 2025
d54d7d8
Use log.exception() instead of log.error(exc_info=True)
dannon Dec 8, 2025
f39451d
Remove unused imports from agent tests
dannon Dec 8, 2025
35ad618
Fix linting issues in API and frontend
dannon Dec 8, 2025
30cff8e
Refactor CustomToolAgent to use UserToolSource directly
dannon Dec 9, 2025
81382dd
Allow local models to attempt structured output, catch specific errors
dannon Dec 9, 2025
7a6c9b8
Fix tool recommendation actions and chat timestamps
dannon Dec 9, 2025
195941c
Use plain string for confidence in tool rec schema to fix vLLM
dannon Dec 9, 2025
84e4d89
Fix router schema for vLLM structured output compatibility
dannon Dec 9, 2025
870e21e
Consolidate AI agent API routes under /api/ai/agents/
dannon Dec 9, 2025
70bcc5d
Use Literal type for confidence to avoid vLLM $defs issue
dannon Dec 9, 2025
ec20298
Fix error wizard visibility for inference_services config
dannon Dec 10, 2025
3f23e4c
Accept encoded job IDs in error-analysis endpoint
dannon Dec 10, 2025
35e6745
Remove hallucinated documentation links from error analysis
dannon Dec 10, 2025
551649b
Formatting cleanup
dannon Dec 10, 2025
67272a3
Add integration tests for AI agent endpoints
dannon Dec 10, 2025
3fa0b7a
Support AI config via GALAXY_TEST_* env vars for integration tests
dannon Dec 10, 2025
ddc1432
Fix suggestions type in error_analysis simple response path
dannon Dec 10, 2025
4fd96da
Add AI tool generation button to CustomToolEditor
dannon Dec 10, 2025
49278bc
Fix structured output detection for gpt-oss models
dannon Dec 10, 2025
7d6b7de
Update test endpoints to use consolidated /api/ai/agents/ routes
dannon Dec 10, 2025
3a2b3e5
Improve model capability error message for custom tool agent
dannon Dec 10, 2025
2be9804
Redesign ChatGXY with simpler, more modern layout
dannon Dec 10, 2025
2e291fd
Remove broken action types and fix SAVE_TOOL to create user tools
dannon Dec 11, 2025
4f3c3b7
Scroll to bottom when loading skeleton appears in ChatGXY
dannon Dec 11, 2025
90fb400
Validate tool availability before recommending in tool agent
dannon Dec 11, 2025
50c48bb
Rebuild client schema
dannon Dec 11, 2025
7279386
Rebuild config/docs
dannon Dec 11, 2025
d548967
Backend formatting/cleanup
dannon Dec 16, 2025
6b658c9
Fix flake8
dannon Dec 11, 2025
8271575
Cleaning up mypy
dannon Dec 16, 2025
96dcec8
More linting fixes, simplified exception handling (per linting rec) a…
dannon Dec 16, 2025
c687efd
Remove unused custom_tool_text.md prompt file
dannon Dec 11, 2025
43bd691
Don't expose exception details to users
dannon Dec 11, 2025
e314497
Rebuild client API schema
dannon Dec 11, 2025
046c98d
Make agent tests more Galactic.
dannon Dec 16, 2025
04043f7
Use AgentResponse schema for agent API endpoints
dannon Dec 11, 2025
fe6e5c2
Fix AgentResponse usage and mypy unused-ignore warnings
dannon Dec 12, 2025
886dac8
Fix test imports for optional pydantic_ai dependency
dannon Dec 12, 2025
d55b7e3
Add unstable parameter for marking experimental API endpoints
dannon Dec 12, 2025
20f5915
Mark AI/chat API endpoints as unstable
dannon Dec 12, 2025
34df7f4
Apply unstable marker to existing experimental endpoints to standardize
dannon Dec 12, 2025
521d8f2
Fix AgentResponse return type in chat.py and regenerate schema
dannon Dec 12, 2025
ee05be1
Remove unused ConfidenceLevel import
dannon Dec 12, 2025
7b9133f
Refactor agent base infrastructure per PR feedback
dannon Dec 12, 2025
358d0b4
Add explicit agent_type to all agent subclasses
dannon Dec 16, 2025
684bb54
Remove tool recommendation agent for separate PR
dannon Dec 18, 2025
541c6b0
Remove more GTN agent references (also extracted separately)
dannon Dec 18, 2025
c66ac4f
Add pydantic-ai as required dependency
dannon Dec 19, 2025
7b8d9ae
Don't expose raw exception details to users in chat API
dannon Dec 19, 2025
ec77f44
Re-solve dependencies with uv
dannon Dec 19, 2025
2cde8d2
Update client api schema.
dannon Dec 19, 2025
c628cd1
Make pydantic-ai a required dependency (remove optional checks)
dannon Dec 19, 2025
7f0775b
Fix mypy type errors in agent and chat code
dannon Dec 19, 2025
b9857d3
Fix _create_agent return types in agent subclasses
dannon Dec 19, 2025
ea116ac
Add type ignore for skipped test with outdated pydantic-ai API
dannon Dec 19, 2025
4b79efd
Add pydantic-ai to galaxy-app package dependencies
dannon Dec 20, 2025
ec6e8a7
Revert mypy and social-auth-core version bumps
dannon Dec 22, 2025
80c1652
Add agents symlink to galaxy-app package
dannon Dec 22, 2025
b73e2e8
Fix mypy errors for optional huggingface imports
dannon Dec 23, 2025
5af2580
Fix mypy error for npz.zip.filelist optional access
dannon Dec 23, 2025
505a93f
Add pytest-asyncio to packages/app test dependencies
dannon Dec 23, 2025
dd0a4a5
fix: failing mocked out integration tests
jmchilton Dec 23, 2025
fc25b0c
Lint fixes.
jmchilton Dec 23, 2025
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
Remove tool recommendation agent for separate PR
Tool-rec code was too entangled across commits to cleanly rebase out,
so just deleting the files and updating references instead.

The agent-based-ai-tool-rec branch has the full implementation for
a follow-up PR.
  • Loading branch information
dannon committed Dec 22, 2025
commit 684bb54016c480ed542b71e31f65978ed85c7dfc
5 changes: 0 additions & 5 deletions 5 client/src/components/ChatGXY.vue
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,6 @@ import {
faThumbsUp,
faTrash,
faUser,
faWrench,
} from "@fortawesome/free-solid-svg-icons";
import { FontAwesomeIcon } from "@fortawesome/vue-fontawesome";
import { BSkeleton } from "bootstrap-vue";
Expand Down Expand Up @@ -83,8 +82,6 @@ interface AgentType {
const agentTypes: AgentType[] = [
{ value: "auto", label: "Auto (Router)", icon: faMagic, description: "Intelligent routing" },
{ value: "error_analysis", label: "Error Analysis", icon: faBug, description: "Debug tool errors" },
{ value: "tool_recommendation", label: "Tool Recommendation", icon: faWrench, description: "Find the right tools" },
{ value: "dspy_tool_recommendation", label: "DSPy Tools", icon: faRobot, description: "Advanced tool selection" },
{ value: "custom_tool", label: "Custom Tool", icon: faPlus, description: "Create custom tools" },
{ value: "dataset_analyzer", label: "Dataset Analyzer", icon: faChartBar, description: "Analyze datasets" },
{ value: "gtn_training", label: "Training Materials", icon: faGraduationCap, description: "Find tutorials" },
Expand All @@ -95,8 +92,6 @@ const agentIconMap: Record<string, IconDefinition> = {
auto: faMagic,
router: faRoute,
error_analysis: faBug,
tool_recommendation: faWrench,
dspy_tool_recommendation: faRobot,
custom_tool: faPlus,
dataset_analyzer: faChartBar,
gtn_training: faGraduationCap,
Expand Down
6 changes: 0 additions & 6 deletions 6 lib/galaxy/agents/__init__.py
Original file line number Diff line number Diff line change
Expand Up @@ -11,12 +11,10 @@
GalaxyAgentDependencies,
)
from .custom_tool import CustomToolAgent
from .dspy_agent import DSPyGalaxyAgent
from .error_analysis import ErrorAnalysisAgent
from .orchestrator import WorkflowOrchestratorAgent
from .registry import AgentRegistry
from .router import QueryRouterAgent
from .tools import ToolRecommendationAgent

__all__ = [
"AgentType",
Expand All @@ -25,10 +23,8 @@
"AgentRegistry",
"QueryRouterAgent",
"ErrorAnalysisAgent",
"ToolRecommendationAgent",
"CustomToolAgent",
"WorkflowOrchestratorAgent",
"DSPyGalaxyAgent",
]

# Global agent registry instance
Expand All @@ -37,7 +33,5 @@
# Register default agents
agent_registry.register(AgentType.ROUTER, QueryRouterAgent)
agent_registry.register(AgentType.ERROR_ANALYSIS, ErrorAnalysisAgent)
agent_registry.register(AgentType.TOOL_RECOMMENDATION, ToolRecommendationAgent)
agent_registry.register(AgentType.CUSTOM_TOOL, CustomToolAgent)
agent_registry.register(AgentType.ORCHESTRATOR, WorkflowOrchestratorAgent)
agent_registry.register(AgentType.DSPY_TOOL_RECOMMENDATION, DSPyGalaxyAgent)
8 changes: 3 additions & 5 deletions 8 lib/galaxy/agents/base.py
Original file line number Diff line number Diff line change
Expand Up @@ -93,11 +93,9 @@ class AgentType:

ROUTER = "router"
ERROR_ANALYSIS = "error_analysis"
TOOL_RECOMMENDATION = "tool_recommendation"
CUSTOM_TOOL = "custom_tool"
GTN_TRAINING = "gtn_training"
ORCHESTRATOR = "orchestrator"
DSPY_TOOL_RECOMMENDATION = "dspy_tool_recommendation"


# Internal agent response model (simplified for internal use)
Expand Down Expand Up @@ -539,7 +537,7 @@ async def _call_agent_from_tool(
reducing code duplication and providing consistent error handling.

Args:
agent_type: Type of agent to call (e.g., "tool_recommendation", "gtn_training")
agent_type: Type of agent to call (e.g., "error_analysis", "gtn_training")
query: Query to send to the target agent
ctx: RunContext from the calling tool function
usage: Optional usage tracking object (defaults to ctx.usage)
Expand All @@ -553,8 +551,8 @@ async def _call_agent_from_tool(

Example usage in @agent.tool functions:
response = await self._call_agent_from_tool(
"tool_recommendation",
f"Find alternatives for: {task}",
"error_analysis",
f"Analyze this issue: {task}",
ctx,
context={"conversation_history": history}
)
Expand Down
303 changes: 0 additions & 303 deletions 303 lib/galaxy/agents/dspy_agent.py

This file was deleted.

Loading
Morty Proxy This is a proxified and sanitized view of the page, visit original site.