advisor-orchestrator-worker

Use when a task is too large for one model pass, needs parallel research or generation across many subtasks (like researching a dozen competitors at once), or the user asks to orchestrate multiple models, split work across a model team, run an advisor-worker loop, have a stronger model review the plan while cheap workers execute, or says "too big for one model" or "fan this out". Not for single-file edits or tasks one model handles in one pass.

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Medium

Suggest reviewing before use

Quality

Content

85%Weight 40%Scale 1-5

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

A well-structured orchestration skill: a clear validated multi-step loop, executable bash with safe fallbacks, and clean one-level reference splits. Minor tightening of the budget/degraded-mode prose and de-templating the inline snippet would push it higher.

Suggestions

Tighten the budget paragraph and degraded-mode explanation to their operative rules; the supporting rationale can be trimmed without losing the guidance.

Make the inline worker-dispatch snippet fully self-contained (show the pids=()/wait loop and variable setup) or move the full version into a reference and keep only the canonical form inline.

Consider extracting the per-wave batching and reaping discipline (batches of 3, per-pid wait, dispatch-order reads) into worker-brief.md so the body stays a lean overview of the loop.

Dimension	Reasoning	Score
Conciseness	Dense and lean with no concept-fluff Claude already knows, but a few rationale-heavy passages (the budget-shape paragraph, the degraded-mode explanation) could be tightened without losing guidance.	4 / 5
Actionability	Provides concrete executable bash (worker dispatch, advisor consult via perl alarm) and complete fallback code in references, but the inline dispatch snippet relies on template placeholders ($brief, $out, pids+=()) that need surrounding setup.	4 / 5
Workflow Clarity	The 7-step loop (Frame→Plan→Plan review→Delegate→Verify→Synthesize→Taste pass) has an explicit Verify checkpoint ('run the actual command, read the actual output'), a FIX→redispatch feedback loop, escalation boundaries, and a status board — full validation for this batch/parallel operation.	5 / 5
Progressive Disclosure	Core loop and hot-path dispatch live inline while the heavier formats/fallbacks are split into three one-level-deep references (advisor-consult.md, worker-brief.md, fallbacks.md), each referenced by exact path in the body with no nested references.	5 / 5
	Total	18 / 20 Passed

Description

100%Weight 40%Scale 1-5

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

A strong, third-person description that concretely states what the skill does, gives natural trigger phrases with synonyms, and carves out a distinct niche with an explicit negative boundary.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'parallel research or generation across many subtasks', 'orchestrate multiple models', 'run an advisor-worker loop', 'have a stronger model review the plan while cheap workers execute' — with comprehensive coverage of the orchestration scenarios.	5 / 5
Completeness	Explicitly answers both 'what' (orchestrate a multi-model team, parallel research/generation, advisor-worker loop) and 'when' ('Use when a task is too large for one model pass...'), plus a concrete negative boundary ('Not for single-file edits or tasks one model handles in one pass').	5 / 5
Trigger Term Quality	Includes natural user phrases a person would actually say, with synonyms: 'too big for one model', 'fan this out', 'split work across a model team', 'orchestrate multiple models'.	5 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (multi-model orchestration) and includes an explicit out-of-scope clause ('Not for single-file edits or tasks one model handles in one pass'), minimizing conflict with single-pass skills.	5 / 5
	Total	20 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: Shubhamsaboo/awesome-llm-apps
Path: agent_skills/advisor-orchestrator-worker/SKILL.md
Commit: 813a55e

Reviewed: 1 day 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.