Skill

maestro-cli

Query Maestro/BAR dependency flow data using the mstro CLI tool via bash. USE FOR: subscription health checks, build flow tracing, codeflow status, channel discovery, triggering subscription updates — when MCP tools aren't loaded or when scripting with JSON output and jq. Also use when investigating "is this subscription stale", "what's the latest build", "check backflow status". DO NOT USE FOR: tasks where maestro MCP tools are already available in context (prefer flow-analysis or flow-tracing skills when MCP server is loaded). INVOKES: bash (mstro CLI commands with --json output).

npx claudepluginhub lewing/agent-plugins --plugin dotnet-dnceng

Tool Access

This skill uses the workspace's default tool permissions.

Preview

Query Maestro/BAR data via the `mstro` CLI tool instead of loading MCP tools into context. Same data, same caching, zero context tax — agents discover capabilities progressively through `--help` and `mstro guide`.

SKILL.md

Similar Skills

buildkite-cli

This skill should be used when the user asks to "trigger a build", "check build status", "watch a build", "view build logs", "retry a build", "cancel a build", "list builds", "download artifacts", "upload artifacts", "manage secrets", "create a pipeline", "list pipelines", or "interact with Buildkite from the command line". Also use when the user mentions bk commands, bk build, bk job, bk pipeline, bk secret, bk artifact, bk cluster, bk package, bk auth, bk configure, bk use, bk init, bk api, or asks about Buildkite CLI installation, terminal-based Buildkite workflows, or command-line CI/CD operations.

4 files

buildkite

maven-build-lifecycle

129

Guides Maven build lifecycles (default, clean, site), phases, goals, profiles, and customization via XML for Java projects. Includes common commands and bindings.

6 tools

maven

using-superpowers

178.4k

Mandates invoking relevant skills via tools before any response in coding sessions. Covers access, priorities, and adaptations for Claude Code, Copilot CLI, Gemini CLI.

3 files

superpowers

Stats

Parent Repo Stars3

Parent Repo Forks1

Last CommitMar 13, 2026

Actions

View Source View Plugin View on GitHub View README

Help us improve

Share bugs, ideas, or general feedback.

Maestro CLI

Query Maestro/BAR data via the mstro CLI tool instead of loading MCP tools into context. Same data, same caching, zero context tax — agents discover capabilities progressively through --help and mstro guide.

When to Use This Skill

Use this skill when:

You need Maestro/BAR data but the maestro MCP server isn't configured
You're scripting or chaining commands with jq, grep, or other CLI tools
You want structured JSON output for downstream processing
The task is one-shot (not a multi-turn conversational investigation)

Prefer MCP-based skills (flow-analysis, flow-tracing) when:

The maestro MCP server is already loaded in your tool list
You're doing multi-turn conversational investigation
You need markdown-formatted output with visual indicators

Installation

dotnet tool install -g lewing.maestro.mcp

After installation, mstro is available globally. Verify: mstro --help

Authentication

Three-tier cascade (automatic fallback):

MAESTRO_BAR_TOKEN env var — explicit PAT
Cached Entra ID — reuses darc authenticate credentials from ~/.darc/.auth-record-*
Anonymous — read-only fallback (may be rate-limited)

Run darc authenticate once for persistent credentials.

Progressive Discovery

Don't memorize commands. Discover them:

mstro --help              # All commands, one line each
mstro <command> --help    # Parameters for a specific command
mstro <command> --schema  # JSON response field names (for jq pipelines)
mstro guide               # Workflow-organized guide (~3KB)

Before writing jq queries, run --schema to see the response shape:

mstro subscription-health --schema   # → shows StaleSubs[].Id, .BuildsBehind, etc.
mstro latest-build --schema          # → shows Id, Repository, Commit, etc.

Common Patterns

All query commands support --json (structured output) and --no-cache (fresh data).

Check subscription health

mstro subscription-health --target-repository https://github.com/dotnet/dotnet --json

Find latest build

mstro latest-build https://github.com/dotnet/runtime --channel-name ".NET 10.0.1xx SDK" --json

Check codeflow status

mstro codeflow-statuses --json

Trace a build graph

mstro build-graph <build-id> --json

Trigger a stale subscription

# Provide --source-repository + --channel-name (auto-resolves latest build)
mstro trigger-subscription <guid> --source-repository https://github.com/dotnet/runtime --channel-name ".NET 10.0.1xx SDK"
# Or provide --build-id directly. Add --force to overwrite stale PR branch.

Chaining with jq

Use mstro <command> --schema to see all response fields before writing jq queries.

# Count stale subscriptions (StaleSubs[].Id, .BuildsBehind, .SourceRepository, .ChannelName)
mstro subscription-health --target-repository https://github.com/dotnet/dotnet --json | jq '.StaleSubs | length'

# Filter channels by name
mstro channels --json | jq '.[] | select(.Name | contains("10.0"))'

# Get build ID then trace graph
BUILD_ID=$(mstro latest-build https://github.com/dotnet/runtime --json | jq -r '.Id')
mstro build-graph $BUILD_ID --json

Cache

Shared SQLite cache at ~/.mstro/cache.db (WAL mode). Cache is shared between CLI and MCP server instances — using the CLI warms the cache for MCP and vice versa.

mstro cache status — show cache stats
mstro cache clear — clear all cached data
--no-cache on any command bypasses cache