app-builder
Main application building orchestrator. Creates full-stack applications from natural language requests. Determines project type, selects tech stack, coordinates agents.
41
27%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/antigravity-app-builder/SKILL.mdQuality
Discovery
32%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 conveys the high-level purpose of the skill (building full-stack apps) and mentions some orchestration steps, but it reads more like an internal system description than a skill selection guide. It lacks explicit trigger guidance ('Use when...'), misses natural user keywords, and uses internal jargon like 'orchestrator' and 'coordinates agents' that users wouldn't say.
Suggestions
Add an explicit 'Use when...' clause with natural trigger phrases, e.g., 'Use when the user asks to build an app, create a new project, scaffold a web application, or generate a full-stack codebase.'
Replace internal jargon ('orchestrator', 'coordinates agents') with user-facing language and include common variations like 'web app', 'new project', 'build', 'create', 'scaffold', 'starter app'.
List more concrete capabilities users would recognize, such as 'generates project structure, configures build tools, sets up frontend and backend, creates boilerplate code.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (full-stack application building) and some actions (determines project type, selects tech stack, coordinates agents), but the actions are more about orchestration mechanics than concrete user-facing capabilities. 'Coordinates agents' is vague. | 2 / 3 |
Completeness | Describes what it does (creates full-stack applications, determines project type, selects tech stack) but completely lacks a 'Use when...' clause or any explicit trigger guidance. Per the rubric, a missing 'Use when...' clause caps completeness at 2, and the 'when' is not even implied clearly, so this scores at 1. | 1 / 3 |
Trigger Term Quality | Includes some relevant terms like 'full-stack applications' and 'natural language requests,' but misses common user phrases like 'build an app,' 'create a website,' 'new project,' 'web app,' 'scaffold,' etc. Users wouldn't naturally say 'orchestrator' or 'coordinates agents.' | 2 / 3 |
Distinctiveness Conflict Risk | The 'full-stack applications' and 'orchestrator' framing provides some distinctiveness, but 'creates applications from natural language requests' is broad enough to overlap with many code-generation or project-scaffolding skills. | 2 / 3 |
Total | 7 / 12 Passed |
Implementation
22%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill functions primarily as a table of contents rather than an actionable orchestration guide. While the navigation structure and template organization are well-presented, the skill lacks any concrete, executable guidance for its core purpose—determining project type, selecting tech stack, and coordinating agents. The usage example is illustrative but not instructional, leaving Claude without clear decision-making criteria or step-by-step procedures.
Suggestions
Add a concrete decision-making workflow with explicit steps: e.g., 'Step 1: Identify keywords from user request using project-detection.md matrix → Step 2: Match to template → Step 3: If ambiguous, ask user to clarify between X and Y'
Include at least a minimal inline decision tree or flowchart for project type detection so the skill is useful even without reading sub-files
Add validation checkpoints to the workflow: e.g., 'Confirm tech stack choice with user before scaffolding', 'Verify all dependencies resolve before coordinating agents'
Replace the vague usage example with a concrete, step-by-step walkthrough showing actual commands or agent invocations that would be executed
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is mostly efficient with good use of tables for navigation, but includes some unnecessary sections like 'When to Use' and 'Limitations' that are generic boilerplate adding no skill-specific value. The selective reading rule header is useful but slightly over-formatted. | 2 / 3 |
Actionability | The skill provides no executable code, no concrete commands, and no specific instructions for how to actually build an application. The usage example is a high-level description of a process rather than actionable steps. It's essentially a table of contents pointing to other files with no standalone guidance. | 1 / 3 |
Workflow Clarity | The usage example shows a numbered list but it's purely descriptive ('Create plan', 'Coordinate agents', 'Report progress') with no concrete steps, no validation checkpoints, no error handling, and no feedback loops. There's no guidance on what to do when detection is ambiguous or when agent coordination fails. | 1 / 3 |
Progressive Disclosure | The file has good structure with clear tables pointing to sub-files and templates with one-level-deep references. However, since no bundle files were provided, we can't verify the referenced paths exist. The SKILL.md itself is almost entirely navigation with virtually no substantive content at the top level, making it too thin as an overview. | 2 / 3 |
Total | 6 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- boisenoise/skills-collections
- Commit
431bfad
- 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.