Jon Saad-FalconandClaude Opus 4.6 c775d620af Simplify model resolution and refactor tool discovery via MCPServer ...

- Remove --router CLI flag from `jarvis ask`; use config fallback chain
  (default_model -> first engine model -> fallback_model) instead
- Simplify SDK _resolve_model() to match the same fallback chain
- Reclassify OrchestratorSFTPolicy and OrchestratorGRPOPolicy as
  IntelligenceLearningPolicy (they fine-tune LM weights, not agent logic)
- Add MCPServer.get_tools() for SystemBuilder integration
- Refactor SystemBuilder._resolve_tools() to discover tools via MCPServer
  with dependency injection, replacing manual per-tool-name imports
- Update CLI router tests to test the new fallback chain

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

2026-02-23 18:41:07 +00:00

..

agents

Add orchestrator training, channels, LiteLLM engine, and simplify learning taxonomy

2026-02-23 18:32:32 +00:00

bench

Implement OpenJarvis v1.0 — all five pillars, SDK, benchmarks, Docker

2026-02-17 00:52:48 +00:00

channels

Add orchestrator training, channels, LiteLLM engine, and simplify learning taxonomy

2026-02-23 18:32:32 +00:00

cli

Simplify model resolution and refactor tool discovery via MCPServer

2026-02-23 18:41:07 +00:00

core

Add orchestrator training, channels, LiteLLM engine, and simplify learning taxonomy

2026-02-23 18:32:32 +00:00

engine

Add orchestrator training, channels, LiteLLM engine, and simplify learning taxonomy

2026-02-23 18:32:32 +00:00

hardware

Expand test suite to 1031 tests: new agents, tools, MCP layer, model catalog

2026-02-21 04:59:11 +00:00

intelligence

Restructure codebase into 5-pillar architecture with MCP tool management, composition layer, and structured learning

2026-02-22 05:57:13 +00:00

learning

Add orchestrator training, channels, LiteLLM engine, and simplify learning taxonomy

2026-02-23 18:32:32 +00:00

mcp

Restructure codebase into 5-pillar architecture with MCP tool management, composition layer, and structured learning

2026-02-22 05:57:13 +00:00

memory

Restructure codebase into 5-pillar architecture with MCP tool management, composition layer, and structured learning

2026-02-22 05:57:13 +00:00

sdk

Implement OpenJarvis v1.0 — all five pillars, SDK, benchmarks, Docker

2026-02-17 00:52:48 +00:00

security

Add 30 tests for security/channels coverage gaps and 6 new docs pages

2026-02-22 03:31:19 +00:00

server

Add orchestrator training, channels, LiteLLM engine, and simplify learning taxonomy

2026-02-23 18:32:32 +00:00

telemetry

Restructure codebase into 5-pillar architecture with MCP tool management, composition layer, and structured learning

2026-02-22 05:57:13 +00:00

test_orchestrator_learning

Add orchestrator training, channels, LiteLLM engine, and simplify learning taxonomy

2026-02-23 18:32:32 +00:00

tools

Add orchestrator training, channels, LiteLLM engine, and simplify learning taxonomy

2026-02-23 18:32:32 +00:00

traces

Expand test suite to 1031 tests: new agents, tools, MCP layer, model catalog

2026-02-21 04:59:11 +00:00

__init__.py

Implement OpenJarvis v1.0 — all five pillars, SDK, benchmarks, Docker

2026-02-17 00:52:48 +00:00

conftest.py

Fix URL summarization, large text handling, and streaming pipeline bugs

2026-02-22 03:17:05 +00:00

test_docker.py

Implement OpenJarvis v1.0 — all five pillars, SDK, benchmarks, Docker

2026-02-17 00:52:48 +00:00

test_integration_extended.py

Restructure codebase into 5-pillar architecture with MCP tool management, composition layer, and structured learning

2026-02-22 05:57:13 +00:00

test_integration.py

Add orchestrator training, channels, LiteLLM engine, and simplify learning taxonomy

2026-02-23 18:32:32 +00:00

test_system.py

Restructure codebase into 5-pillar architecture with MCP tool management, composition layer, and structured learning

2026-02-22 05:57:13 +00:00