task-decomposer
Decomposes complex user requests into executable subtasks, identifies required capabilities, searches for existing skills at skills.sh, and creates new skills when no solution exists. This skill should be used when the user submits a complex multi-step request, wants to automate workflows, or needs help breaking down large tasks into manageable pieces.
73
47%
Does it follow best practices?
Impact
90%
2.43xAverage score across 6 eval scenarios
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./public/skills/10e9928a/task-decomposer/SKILL.mdQuality
Discovery
67%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 structurally complete with both 'what' and 'when' clauses clearly present, which is its main strength. However, the actions described are abstract and meta-level rather than concrete, and the trigger terms are somewhat generic, creating potential overlap with many other skills. The description would benefit from more specific, natural trigger terms and clearer differentiation from other task-management or planning skills.
Suggestions
Add more natural trigger terms users would actually say, such as 'plan this out', 'step by step', 'orchestrate', 'find a skill for', or 'I need to do multiple things'.
Increase distinctiveness by explicitly naming what makes this different from general task-planning skills—e.g., mention the specific integration with skills.sh and the skill creation workflow as unique differentiators.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names several actions—decomposing requests, identifying capabilities, searching for skills, creating new skills—but they are somewhat abstract and process-oriented rather than concrete end-user-visible actions. It describes a meta-process rather than specific tangible outputs. | 2 / 3 |
Completeness | The description clearly answers both 'what' (decomposes requests, identifies capabilities, searches for skills, creates new skills) and 'when' with an explicit trigger clause ('should be used when the user submits a complex multi-step request, wants to automate workflows, or needs help breaking down large tasks'). | 3 / 3 |
Trigger Term Quality | Includes some relevant terms like 'complex multi-step request', 'automate workflows', and 'breaking down large tasks', but these are somewhat generic. Users might say 'break this into steps' or 'plan this out' but the description lacks those more natural phrasings and common variations. | 2 / 3 |
Distinctiveness Conflict Risk | The skill is meta-level (orchestrating other skills), which gives it some distinctiveness, but terms like 'complex requests', 'automate workflows', and 'large tasks' are quite broad and could easily overlap with many other skills that also handle complex or multi-step tasks. | 2 / 3 |
Total | 9 / 12 Passed |
Implementation
27%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is significantly over-engineered and verbose for what it does. It spends most of its token budget on conceptual frameworks (capability taxonomies, decomposition principles, ASCII output templates) that Claude can generate on its own, rather than focusing on the few actionable pieces: how to search skills.sh, how to install skills, and when to create new ones. The content would benefit enormously from aggressive trimming and splitting detailed reference material into separate files.
Suggestions
Reduce the content by 70%+: remove the Universal Capability Types table, the Task Decomposition Principles section, the ASCII output format template, and the detailed skill creation template — Claude already knows how to do all of these things.
Split remaining reference material (e.g., skill template, example decompositions) into separate files like EXAMPLES.md and SKILL-TEMPLATE.md, keeping SKILL.md as a concise overview.
Add explicit validation checkpoints: 'Verify skill installation with npx skills list before proceeding' and 'If no skills found, try broader/narrower search terms before creating new skills'.
Focus the main file on the 3-4 concrete, unique pieces of knowledge: the npx skills CLI commands, the skills.sh URL, the decision tree for search-vs-create, and user confirmation requirements.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at ~350+ lines. The 'Universal Capability Types' table, ASCII art output format, detailed YAML templates, and extensive examples all explain things Claude already knows how to do. The skill could be reduced to ~25% of its size without losing actionable information. | 1 / 3 |
Actionability | Provides concrete CLI commands (npx skills find, npx skills add) and structured YAML templates, but most content is conceptual framework and taxonomy rather than executable guidance. The YAML decomposition examples are illustrative templates, not copy-paste-ready code. | 2 / 3 |
Workflow Clarity | The 6-phase workflow is clearly sequenced and logically ordered, but lacks explicit validation checkpoints and error recovery loops. There's no 'if search fails, try alternative keywords' or 'verify skill installation succeeded before proceeding' feedback mechanism built into the steps. | 2 / 3 |
Progressive Disclosure | Monolithic wall of text with everything inline. The capability taxonomy table, skill template, output format template, and multiple full examples could all be split into separate reference files. No references to external files for detailed content. | 1 / 3 |
Total | 6 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- Demerzels-lab/elsamultiskillagent
- Commit
45f9fac
- Reviewed
Table of Contents
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.