minion-orchestrator

Unified Minions skill for both deterministic shell jobs and LLM subagent orchestration. Replaces the older `gbrain-jobs` routing intent. Use when: submitting gbrain jobs, shell/background tasks, spawning subagents, checking progress, steering running work, pausing/resuming, parallel fan-out. One durable, observable, steerable queue interface.

Quality

96%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Quality

Content

92%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable and concise with a well-sequenced, validated workflow, but it is monolithic — substantial reference material (tool allow-lists, CLI flag tables) is inline with no bundle files or one-level-deep navigation to split it out.

Suggestions

Move the BRAIN_TOOL_ALLOWLIST and the `gbrain jobs submit --help` flag tables into a reference file (e.g. references/cli-reference.md) and link to it, keeping the body a concise overview.

Add a short 'References' section pointing to the referenced project files (subagent-routing.md, minions-shell-jobs.md, handler sources) so the inline pointers become navigable one-level-deep links.

Split the detailed Phase 1–5 subagent lifecycle into a separate guide and summarize the phases in the body, reducing inline bulk while preserving the workflow sequence.

Dimension	Reasoning	Score
Conciseness	Lean throughout — tables, code blocks, and concrete commands with no exposition of concepts Claude already knows; every section earns its place for an orchestrator skill. Minor padding in the Guarantees/flag lists does not drop it to anchor 2.	3 / 3
Actionability	Provides fully executable, copy-paste-ready commands — `gbrain jobs submit shell --params '{...}'`, `gbrain agent run "..." --tools "search,query"`, and MCP calls with exact parameter syntax like `replay_job id=ID data_overrides={...}`.	3 / 3
Workflow Clarity	Clear five-phase sequence (Submit → Monitor → Steer → Lifecycle → Review) with a Preconditions gate, an explicit verification step ("run `gbrain jobs stats` to confirm the worker is registered"), and guardrails for batch work ("Don't spawn > 5 concurrent agents without checking stats first").	3 / 3
Progressive Disclosure	No reference/scripts/assets bundle exists, so all detail lives inline in a single ~260-line SKILL.md; in-file section organization is good but full allow-lists and flag tables that could be split into reference files are kept inline, matching the 'content that should be separate is inline' anchor.	2 / 3
	Total	11 / 12 Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific, trigger-rich, and complete, naming concrete capabilities and an explicit 'Use when' clause with natural user phrasing in third-person voice. It is clearly distinct from other skills.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "submitting gbrain jobs, shell/background tasks, spawning subagents, checking progress, steering running work, pausing/resuming, parallel fan-out" — matching the 'multiple specific concrete actions' anchor.	3 / 3
Completeness	Clearly states what the skill does ("durable, observable, steerable queue interface" for shell + subagent jobs) and provides an explicit "Use when:" clause with multiple triggers, answering both what and when.	3 / 3
Trigger Term Quality	Includes natural phrases users would say ("submit a gbrain job", "shell job", "spawn agent", "run in background", "check on agent", "fan out"), giving good coverage of common variations.	3 / 3
Distinctiveness Conflict Risk	Occupies a clear niche — a Postgres-native job queue with gbrain-specific triggers — making it unlikely to fire for unrelated skills.	3 / 3
	Total	12 / 12 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: garrytan/gbrain
Path: skills/minion-orchestrator/SKILL.md
Commit: 3fafb69

Reviewed: about 19 hours ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.