thaloco/pyro-mcp
1
Fork 0
Code Pull requests Releases Packages Activity Actions

Clarify workspace-core as the chat-host onramp

Browse source 
Make the recommended MCP profile visible from the first help and docs pass without changing 3.x behavior.

Rework  help, top-level docs, public-contract wording, and shipped MCP/OpenAI examples so  is the recommended first profile while  stays the compatibility default for full-surface hosts.

Bump the package and catalog to 3.8.0, mark the roadmap milestone done, and add regression coverage for the new MCP help and docs alignment. Validation included uv lock, targeted profile/help tests, make check, make dist-check, and a real guest-backed  server smoke.
This commit is contained in:
Thales Maciel 2026-03-13 11:23:51 -03:00
parent 7a0620fc0c
commit 407c805ce2
17 changed files with 150 additions and 50 deletions
Download patch file Download diff file Expand all files Collapse all files

6
examples/mcp_client_config.md
View file

@ -1,5 +1,7 @@
# MCP Client Config Example
Recommended default for most chat hosts: `workspace-core`.
`pyro-mcp` is intended to be exposed to LLM clients through the public `pyro` CLI.
Generic stdio MCP configuration using `uvx`:
@ -30,9 +32,9 @@ If `pyro-mcp` is already installed locally, the same server can be configured wi
Profile progression:
- `workspace-core`: the recommended first persistent chat profile
- `vm-run`: expose only `vm_run`
- `workspace-core`: the default persistent chat profile
- `workspace-full`: shells, services, snapshots, secrets, network policy, and disk tools
- `workspace-full`: advanced 3.x compatibility surface for shells, services, snapshots, secrets, network policy, and disk tools
Primary profile for most agents:

7
examples/openai_responses_workspace_core.py
View file

@ -4,9 +4,10 @@ Requirements:
- `pip install openai` or `uv add openai`
- `OPENAI_API_KEY`
This example mirrors the `workspace-core` MCP profile by deriving tool schemas
from `Pyro.create_server(profile="workspace-core")` and dispatching tool calls
back through that same profiled server.
This is the recommended persistent-chat example. It mirrors the
`workspace-core` MCP profile by deriving tool schemas from
`Pyro.create_server(profile="workspace-core")` and dispatching tool calls back
through that same profiled server.
"""
from __future__ import annotations