CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

scaffold-project

Scaffold a new app, API, backend, fullstack project, monorepo or starter with Better-T-Stack, including projects on a named framework such as Hono, Express, Fastify, Elysia, Convex, Next.js, TanStack Router or Start, React Router, Nuxt, Svelte, Solid, Astro or React Native. Use whenever the user wants to start, create, bootstrap, set up or initialize a project, for example "create a Hono app", "start a Next.js project" or "scaffold a fullstack app with auth and a database". Prefer it over hand-writing package.json, configs and folders.

Quality

95%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

100%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.

An exemplary skill body: lean prose with zero padding, fully executable commands and tool calls, a seven-step workflow with explicit plan-fail-fix-retry and verification checkpoints, and clean two-file progressive disclosure verified against the actual bundle. No dimension showed anchor-level deficiencies.

DimensionReasoningScore

Conciseness

The body explains nothing Claude already knows and every rule carries its justification in a single clause ("Installs can outlast tool timeouts, so create with `install: false` and install as a separate step"). All value enumerations and payloads are deferred to references, so every token earns its place — anchor 5, not 4, since no over-explanation to trim was found.

5 / 5

Actionability

Guidance is executable end to end: "npx create-better-t-stack@latest schema --name createInput", "`bts_create_project` with `install: false`", '`create-json` with "dryRun": true', and "Run `<packageManager> install` in the new folder", with the full copy-paste payload in references/cli.md. Anchor 5; the few deliberately generic spots ("typecheck or build script from its package.json") are appropriately flexible rather than gaps.

5 / 5

Workflow Clarity

Seven sequenced steps include an explicit feedback loop ("If the plan fails, its error names the conflict. Fix the config and plan again"), a user checkpoint ("Show the user the planned stack and get a yes"), and post-create verification ("Prove it runs... If a step fails, report the exact error"). This matches anchor 5's explicit validation and error-recovery loop; the skill creates new directories rather than performing destructive/batch edits, so the cap of 3 does not apply.

5 / 5

Progressive Disclosure

A concise overview body splits detail into two one-level-deep references, each clearly signaled with its contents: "[references/options.md] lists the values and the common compatibility rules" and "[references/cli.md] has the full `create-json` payload, the flag form for a command the user pastes, and the Stack Builder URL keys". Both files exist and contain exactly what is claimed, matching anchor 5's well-signaled, appropriately split structure.

5 / 5

Total

20

/

20

Passed

Description

91%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 with an explicit what, an explicit 'Use whenever...' trigger clause packed with natural synonyms and example utterances, and a clear tool-specific niche. Its only weaknesses are a single-verb action list and a 'whenever a project' trigger broad enough to over-fire on unsupported stacks.

Suggestions

Qualify the 'Use whenever' trigger with the ecosystems Better-T-Stack supports (e.g., 'for JavaScript/TypeScript projects') so it does not claim Django, Rails or other unsupported stacks.

Add one or two more distinct concrete actions (e.g., 'generates package.json, configs, auth and database wiring') to round out the action list beyond the single 'scaffold' verb.

DimensionReasoningScore

Specificity

"Scaffold a new app, API, backend, fullstack project, monorepo or starter with Better-T-Stack" and "Prefer it over hand-writing package.json, configs and folders" name concrete actions and artifacts, but the action list centers on one verb across many target types rather than multiple distinct operations. This matches anchor 4 ("several specific actions; minor gaps"), not 5's comprehensive multi-action list and not 3's 1-2 concrete actions, since target-type and framework coverage is extensive.

4 / 5

Completeness

It explicitly answers what ("Scaffold a new app, API, backend, fullstack project, monorepo or starter with Better-T-Stack") and when ("Use whenever the user wants to start, create, bootstrap, set up or initialize a project, for example...") with concrete trigger phrases. This is an exact match to the anchor 5 example pattern; the explicit 'Use when...' clause means the completeness cap of 3 does not apply.

5 / 5

Trigger Term Quality

Natural synonyms "start, create, bootstrap, set up or initialize" plus "scaffold", named frameworks ("Hono, Express, Fastify, Elysia, Convex, Next.js, TanStack Router or Start, React Router, Nuxt, Svelte, Solid, Astro or React Native") and verbatim user utterances ("create a Hono app", "start a Next.js project") give comprehensive natural-term coverage. Anchor 5 fits; 4 would require a missing common variation, and none is evident.

5 / 5

Distinctiveness Conflict Risk

The tool-specific niche ("with Better-T-Stack") and the disambiguator "Prefer it over hand-writing package.json, configs and folders" make it mostly distinct, but "Use whenever the user wants to start, create, bootstrap, set up or initialize a project" could fire for stacks the tool does not support (e.g., Django, Rails). That is minor overlap risk with closely related scaffolding skills — anchor 4 — rather than anchor 5's minimal conflict risk, and clearly above anchor 3's broader overlap.

4 / 5

Total

18

/

20

Passed

Validation

87%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

Total

14

/

16

Passed

Repository

AmanVarshney01/create-better-t-stack

Path

plugin/skills/scaffold-project/SKILL.md

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.