decomposition

Resolves task dependencies, generates machine-actionable delegation specs, structures phased subtask plans for multi-agent work. Use when writing delegation specs, resolving task dependencies, building phased subtask plans for multi-agent work, assigning work to sub-agents, or partitioning a feature into parallelizable phases.

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

82%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-crafted, token-efficient instruction skill: concrete templates, decision trees, and worked examples with almost no padding. The main gaps are unexplained external conventions (complexity score /13, TAS-XX, Foundation References) that reduce self-contained actionability and navigation.

Suggestions

Briefly define or link the complexity-score system ([score]/13 → tier) and TAS-XX tracker convention so the Delegation Spec template is fully self-contained.

List or point to the "5 Foundation References" and the LESSONS-LEARNED.md location so deferred content is actually navigable.

Add an explicit inter-phase validation checkpoint (e.g., what the QA gate must verify before proceeding) to strengthen workflow feedback loops.

Dimension	Reasoning	Score
Conciseness	The body is dense and lean — tables, templates, and a decision tree with no concept-explanation padding ("`TaskB → TaskA` = B depends on A (must finish first)") — and it assumes Claude's competence (complexity scores, TAS-XX) rather than over-explaining, matching anchor 5.	5 / 5
Actionability	The Delegation Spec template is paste-ready and the strong/weak prompt examples and delegation decision tree are concrete and executable, but the "[score]/13 → tier" complexity system and TAS-XX tracker conventions are used without being defined in this file or bundle, leaving minor gaps that keep it at anchor 4 rather than fully self-contained 5.	4 / 5
Workflow Clarity	A coherent pipeline is present (dependency resolution → phased plan → delegation mode selection → QA gate "verify phases, run builds") plus a feedback loop ("Add lesson if you retry any approach"), but inter-phase validation checkpoints and error-recovery steps are implicit rather than explicit, fitting anchor 4.	4 / 5
Progressive Disclosure	Sections are well organized and detail is properly deferred one level deep ("Load project-consistency skill for full Foundation Phase pattern, prompt templates, anti-patterns"), but "the 5 Foundation References" and `.opencastle/LESSONS-LEARNED.md` are invoked without being listed or located, and no bundle files exist to back them — anchor 4, not fully navigable 5.	4 / 5
	Total	17 / 20 Passed

Description

87%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 description: third-person, concrete, and with an explicit multi-trigger "Use when" clause that closely matches the rubric's good examples. Minor room to improve via added trigger synonyms (e.g., "break down", "split up") and slightly richer what-coverage beyond the trigger list.

Suggestions

Add natural synonym triggers such as "break down tasks" or "split up work" to broaden trigger_term_quality.

Differentiate the capability list from the trigger list, e.g., mention cycle detection or parallelization analysis as distinct capabilities.

Dimension	Reasoning	Score
Specificity	The description lists three concrete actions ("Resolves task dependencies, generates machine-actionable delegation specs, structures phased subtask plans for multi-agent work") with good coverage of the domain, but the action list largely mirrors the trigger list and adds little additional capability detail, matching anchor 4 rather than a fully comprehensive 5.	4 / 5
Completeness	It explicitly answers both "what" (three third-person concrete actions) and "when" (a full "Use when..." clause with multiple concrete trigger phrases), matching the anchor-5 example structure; the "when" is fully explicit, not merely present as at anchor 4.	5 / 5
Trigger Term Quality	The "Use when" clause covers natural phrases users would say — "writing delegation specs", "assigning work to sub-agents", "partitioning a feature into parallelizable phases" — but omits common variations such as "break down tasks" or "split up work", fitting anchor 4 rather than comprehensive synonym coverage.	4 / 5
Distinctiveness Conflict Risk	Terms like "machine-actionable delegation specs", "phased subtask plans for multi-agent work", and "sub-agents" carve a clear niche with distinct triggers and minimal conflict risk against generic planning or coding skills.	5 / 5
	Total	18 / 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: monkilabs/opencastle
Path: src/orchestrator/skills/decomposition/SKILL.md
Commit: c0f4566

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.