llama-stack-mirror/llama_stack/providers/inline
Sébastien Han df6ce8befa
fix: only load mcp when enabled in tool_group (#2621)
# What does this PR do?

The agent code is currently importing MCP modules even when MCP isn’t
enabled. Do we consider this worth fixing, or are we treating MCP as a
first-class dependency? I believe we should treat it as such.

If everyone agrees, let’s go ahead and close this.

Note: The current setup breaks if someone builds a distro without
including MCP in tool_group but still serves the agent API.

Also, we should bump the MCP version to support streamable responses, as
SSE is being deprecated.

Signed-off-by: Sébastien Han <seb@redhat.com>
2025-07-04 20:27:05 +05:30
..
agents fix: only load mcp when enabled in tool_group (#2621) 2025-07-04 20:27:05 +05:30
datasetio chore(refact): move paginate_records fn outside of datasetio (#2137) 2025-05-12 10:56:14 -07:00
eval chore: remove nested imports (#2515) 2025-06-26 08:01:05 +05:30
files/localfs refactor(env)!: enhanced environment variable substitution (#2490) 2025-06-26 08:20:08 +05:30
inference refactor(env)!: enhanced environment variable substitution (#2490) 2025-06-26 08:20:08 +05:30
ios/inference chore: removed executorch submodule (#1265) 2025-02-25 21:57:21 -08:00
post_training fix: proper checkpointing logic for HF trainer (#2429) 2025-06-27 17:36:25 -04:00
safety feat: add llama guard 4 model (#2579) 2025-07-03 22:29:04 -07:00
scoring fix: allow default empty vars for conditionals (#2570) 2025-07-01 14:42:05 +02:00
telemetry docs: auto generated documentation for providers (#2543) 2025-06-30 15:13:20 +02:00
tool_runtime feat: Add ChunkMetadata to Chunk (#2497) 2025-06-25 15:55:23 -04:00
vector_io fix: Fixing Milvus sample config and updating documentation (#2568) 2025-06-30 19:25:23 -07:00
__init__.py impls -> inline, adapters -> remote (#381) 2024-11-06 14:54:05 -08:00