appfactory-builder
Build and deploy production apps using AppFactory's 7 pipelines (websites, mobile, dApps, AI agents, plugins, mini apps, bots). One prompt → live URL.
61
51%
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 ./data/skills-md/0xaxiom/appfactory/appfactory-builder/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 strong in specificity and distinctiveness, clearly naming the tool (AppFactory), its 7 pipeline types, and the expected output (live URL). However, it lacks an explicit 'Use when...' clause, which caps completeness, and could benefit from more natural user-facing trigger terms beyond the pipeline type names.
Suggestions
Add an explicit 'Use when...' clause, e.g., 'Use when the user wants to build, deploy, or launch a production app, website, bot, or dApp using AppFactory.'
Include more natural trigger variations users might say, such as 'web app', 'build a site', 'deploy an app', 'create a chatbot', or 'launch a project'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions and outputs: 'Build and deploy production apps' across 7 named pipeline types (websites, mobile, dApps, AI agents, plugins, mini apps, bots), with a clear outcome ('One prompt → live URL'). | 3 / 3 |
Completeness | Clearly answers 'what does this do' (build and deploy apps using AppFactory's 7 pipelines), but lacks an explicit 'Use when...' clause or equivalent trigger guidance for when Claude should select this skill. | 2 / 3 |
Trigger Term Quality | Includes several relevant keywords like 'websites', 'mobile', 'dApps', 'AI agents', 'plugins', 'mini apps', 'bots', and 'deploy', but misses common user phrasings like 'web app', 'build a site', 'launch', 'host', or 'AppFactory' as an explicit trigger. The term 'AppFactory' appears but may not be what users naturally say. | 2 / 3 |
Distinctiveness Conflict Risk | The mention of 'AppFactory's 7 pipelines' and the specific list of pipeline types (dApps, mini apps, bots, etc.) creates a clear niche that is unlikely to conflict with generic coding or deployment skills. | 3 / 3 |
Total | 10 / 12 Passed |
Implementation
35%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides a reasonable overview of AppFactory's pipelines with some concrete commands for the website pipeline, but is significantly bloated with promotional content, contribution guidelines, architecture details, and token/social media links that don't help Claude build apps. The actionable content is diluted by marketing material, and most pipelines lack inline executable guidance, deferring to external CLAUDE.md files without summarizing key steps. Validation and error handling are absent from the workflow.
Suggestions
Remove all promotional content (token links, social media handles, showcase submission instructions, contributing guidelines) — these don't help Claude execute builds and waste significant token budget.
For each pipeline beyond websites, include at least the 3-5 key commands inline rather than just 'Follow CLAUDE.md' — summarize the critical path so Claude can act without reading another file.
Add explicit validation checkpoints to the workflow: verify `npm run build` succeeds before deploying, check for the LOCAL_RUN_PROOF_GATE mentioned in architecture, and include error recovery steps.
Move Architecture, Design System, and Contributing sections to separate referenced files, keeping SKILL.md focused on the quick-start build-and-deploy workflow.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~200+ lines. It includes extensive sections on contributing, reporting issues, architecture details, OpenClaw agent heartbeat tasks, token launchpad links, and social media handles — none of which help Claude execute the core task. The pipeline details section largely repeats information from the table. Marketing language ('Ship something') and promotional content ($FACTORY token links) waste tokens. | 1 / 3 |
Actionability | The Quick Start section provides concrete bash commands for cloning, scaffolding, and deploying, which is good. However, most pipelines just say 'Follow <pipeline>/CLAUDE.md' without providing the actual executable steps. The website pipeline is the only one with inline commands; the rest defer entirely to external files without summarizing key steps. | 2 / 3 |
Workflow Clarity | There's a numbered sequence (clone → pick pipeline → build → deploy) which provides basic structure. However, there are no validation checkpoints — no step to verify the build succeeded before deploying, no error recovery guidance, and the LOCAL_RUN_PROOF_GATE is mentioned in the architecture section but never integrated into the workflow steps. For a multi-step build-and-deploy process, missing validation caps this at 2. | 2 / 3 |
Progressive Disclosure | The skill references per-pipeline CLAUDE.md files for detailed instructions, which is good progressive disclosure. However, the main file itself is bloated with sections that should be in separate files (Contributing, Architecture, OpenClaw Agents, Design System) or omitted entirely. The inline content that should be split out undermines the structure. | 2 / 3 |
Total | 7 / 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
- NeverSight/skills_feed
- Commit
e437c3c
- 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.