mco

MCO v0.12.1 adds Antigravity CLI support and publishes the Windows, MCP, and Ollama reliability fixes merged since v0.11.0.

Added

• Built-in Antigravity CLI (agy) adapter with model selection, execution modes, authentication checks, and permission-risk reporting (#127).
• MCP progress notifications during long runs and per-call timeout overrides (#125).

Fixed

• Install locked Node dependencies before the release tests. The initial v0.12.0 release attempt failed before npm publication; v0.12.1 includes the same product changes with the corrected publish workflow.
• Windows npm/Python command-shim execution and Python runtime compatibility (#123, #124).
• MCP calls now honor configured timeouts, whole-run deadlines, parallelism limits, and pinned models (#125).
• Provider answers survive legacy console encodings instead of disappearing (#125).
• Ollama readiness checks resolve the exact requested model and treat flag-like names as positional arguments (#128, #129).

Update

npm install -g @tt-a1i/[email protected]
mco --version

Full changelog: https://github.com/mco-org/mco/compare/v0.11.0...v0.12.1

v0.11.0 - Invocation-native runtime

Status: breaking release guidance

This release changes MCO from a findings/decision pipeline into a raw-answer invocation runtime. Legacy findings-driven design documents remain available for context but are superseded by the invocation runtime v1 contract; this note is the migration guide for callers adopting the new public surface.

New behavior

• Use repeatable --agent [alias=]provider:model for explicit Agent identity and multiple models from one Provider.
• Use --providers as the invocation-native shorthand for one configured/default model per Provider.
• Consume raw answer text plus operational invocation records. Task status is complete, partial, or failed, with exit codes 0, 1, and 2.
• Use text mode, --json, or --stream jsonl; stdout contains answers or the selected machine protocol, while diagnostics go to stderr.
• Use --chain, --debate, and --synthesize for LLM-driven file-backed stages.
• Use --result-mode artifact|stdout|both for temporary or persistent Markdown artifacts.
• A synthesis manifest retains run and debate records, including failures or missing entries, but never includes synthesis itself. Root result.md groups actual stages, places synthesis first, and keeps every original answer and failure record.

Breaking removals

The following surfaces are removed:

• mco findings and the findings JSON schema;
• semantic normalization, deduplication, confidence, consensus, passive lifecycle, and findings-driven memory runtime;
• Markdown-PR and SARIF renderers;
• content-based INCONCLUSIVE status;
• --format, --strict-contract, --memory, --space, --diff, --staged, --unstaged, and --diff-base.

Removed flags return migration guidance rather than being silently ignored.

Migration examples

Old structured review callers should send the desired scope and output shape as a raw prompt:

mco review --target-paths runtime --providers claude,codex \
  --prompt "Review runtime/ for correctness and security risks; report your reasoning in natural language." \
  --result-mode both --task-id runtime-review

Old multi-model callers should use explicit invocations:

mco run --prompt "Compare these implementations." \
  --agent fast=pi:model-a \
  --agent careful=pi:model-b \
  --stream jsonl

Old persistent-memory callers should save and pass Markdown artifacts through chain/debate/synthesis stages:

mco run --prompt "Analyze the repository." \
  --providers claude,codex --chain --execution-mode read_only \
  --result-mode artifact --task-id repository-analysis

Write safety

MCO does not create or manage worktrees. If a user explicitly chooses parallel writers, the calling Agent must partition non-overlapping --target-paths and warn about overlapping edits.

What's Changed

• docs: add npm release guide by @tt-a1i in https://github.com/mco-org/mco/pull/96
• feat: add GitHub Copilot CLI adapter by @anxkhn in https://github.com/mco-org/mco/pull/98
• feat: ship agent-native ergonomics in v0.10.7 by @tt-a1i in https://github.com/mco-org/mco/pull/100
• feat: add explicit agent selection and execution modes by @tt-a1i in https://github.com/mco-org/mco/pull/101
• feat: add installer and skill sync workflow by @tt-a1i in https://github.com/mco-org/mco/pull/102

New Contributors

• @anxkhn made their first contribution in https://github.com/mco-org/mco/pull/98

Full Changelog: https://github.com/mco-org/mco/compare/v0.10.6...v0.10.8

Fixed

• Concatenate multi-part JSONL text_delta streams in event order for run-mode final text extraction.
• Pass --dir <repo_root> to OpenCode so review/run tasks inspect the selected repository.

Changed

• Clarify that MCO defaults to the audited five providers while supporting seven providers explicitly.
• Mark Hermes as an explicit elevated opt-in provider because Hermes oneshot mode auto-bypasses approvals.
• Expand manual capability probes and step5 benchmark defaults to include Hermes and Pi.

Validation

• PYTHONDONTWRITEBYTECODE=1 python3 -m unittest discover -s tests -p 'test_*.py' -> 696 tests OK
• GitHub Gate on PR #89 -> success
• npm pack --dry-run --json -> 0.10.3 pack OK
• local npm install smoke -> Pi run returned OK
• live 5-provider run/review smoke -> claude/codex/opencode/hermes/pi all succeeded; review parse/schema valid for all 5