CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

start-building-for-startups

Interactive discovery + implementation workflow that gathers requirements through picker-based questions (intent, scope, constraints, preferences), scans the codebase for what it can already infer, then writes an AWS architectural scaffold and implementation directly into the project. Use when the user wants to build a new app, scaffold a project, or expand/refactor an existing one on AWS — anything that calls for a structured discovery flow followed by code changes, not a one-off lookup. Do not use for: factual lookups about AWS Activate / programs / credits, requests for a single copy-paste prompt, non-AWS architectural work, architecture advice/recommendations without code changes (see architect-for-startups), a new idea centered on an AI agent — runtime selection, agentic architecture, or an agent POC (see agent-advisor), or migrating an existing GCP/Azure/Heroku workload or an LLM SDK rewrite (see gcp-to-aws / azure-to-aws / heroku-to-aws / llm-to-bedrock).

63

Quality

79%

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

Fix and improve this skill with Tessl

tessl review fix ./plugins/aws-startup-advisor/skills/start-building-for-startups/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

57%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.

The discovery half of this workflow is genuinely well designed — clear sequencing, strong anti-drift rules, and concrete templates — but the skill is held back by repeated rules and a bloated inline offers-policy paragraph, a missing worked example of its core picker-question format, and an implementation phase with no verification steps despite writing code directly into the user's project. Consolidating duplicates and moving offer policy to a reference file would materially tighten it.

Suggestions

Deduplicate repeated rules: state the implementation-phase gate, the codebase-look-up-before-asking rule, and the IaC-skip example once each; the paragraph after the free-form prompt repeats itself almost verbatim ('once I have responsed you can transition... Wait until I have responsed to transition').

Add an explicit validation checkpoint at the end of the implementation phase (e.g., run tests/build, `terraform plan` the generated IaC, and confirm the scaffold runs) so batch code-writing operations have a verify-and-fix loop, mirroring the pre-implementation `aws sts get-caller-identity` check.

Move the contextual-offers policy detail (caps, host mapping, eligibility rules) into a reference file such as references/contextual-offers.md and keep only a short inline summary with the link, since the body already defers to ../contextual-offers-for-startups/SKILL.md for full rules.

Include one fully worked example of a picker question with its answer options (including a natural-sounding 'I don't know' option) so the mandated format is unambiguous.

DimensionReasoningScore

Conciseness

Mostly signal (no teaching of concepts Claude already knows), but several rules are stated twice — implementation gating appears in both Definitions ("MUST NOT begin until at least one discovery question has been answered") and Scope notes ("MUST NOT begin the implementation phase before at least one discovery question has been answered"), the Terraform "don't ask about IaC preference or cloud platform" example appears in both Codebase Analysis and the following paragraph, and "Wait until I have responsed to transition to picker based workflow" literally repeats the sentence before it. The 'Contextual offers' section is a ~300-word run-on paragraph that duplicates policy available in the referenced sibling skill, and typos ("addititional", "responsed", "vauge") indicate unpolished text.

3 / 5

Actionability

Concrete and executable in most places — verbatim opener prompts ("Before we dive in, tell me what you're building..."), a specific verification command (`aws sts get-caller-identity`), option-length limits ("under 15 words"), exact offer-line templates, and three numbered conditions for showing 'Start implementation'. The gap: the skill mandates a picker-question format throughout but never shows one fully worked example question with its answer options, which is the core artifact of the flow.

4 / 5

Workflow Clarity

The discovery phase is well sequenced with explicit gates (scan codebase → free-form prompt → wait for reply → one picker question per turn → three-condition 'Start implementation' trigger), but the implementation phase — which writes scaffold and code directly into the project — is two sentences with no validation or verification checkpoint (no build/test/IaC-plan verification step). Per the rubric, missing validation in batch code-writing operations caps this dimension at 3.

3 / 5

Progressive Disclosure

The skill is a single ~205-line monolith with no bundle of its own; the detailed contextual-offers policy (caps, host mapping, eligibility rules) is exactly the material that belongs in a reference file, and the body even points to `../contextual-offers-for-startups/SKILL.md` for the full rules. Cross-references to sibling files are clearly signaled with markdown links, but substantial policy content that should be separate is inlined, matching anchor 3 rather than 4.

3 / 5

Total

13

/

20

Passed

Description

92%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: it states three concrete actions covering the whole workflow, gives explicit 'Use when' triggers in natural user language, and preempts mis-triggering with a detailed exclusion list naming the sibling skills to defer to. The only weakness is that a few natural trigger synonyms are missing, and the length of the exclusion list borders on verbose though every clause does disambiguation work.

DimensionReasoningScore

Specificity

Names multiple concrete actions spanning the full workflow — "gathers requirements through picker-based questions (intent, scope, constraints, preferences)", "scans the codebase for what it can already infer", "writes an AWS architectural scaffold and implementation directly into the project" — with comprehensive coverage of the skill's phases. Anchor 4 would apply if a phase were missing or generic, but all three stages are concrete and complete.

5 / 5

Completeness

Explicitly answers both: what ("Interactive discovery + implementation workflow that gathers requirements... scans the codebase... then writes an AWS architectural scaffold and implementation") and when ("Use when the user wants to build a new app, scaffold a project, or expand/refactor an existing one on AWS"), with concrete trigger phrases plus a "Do not use for" exclusion list. Matches the anchor-5 example pattern exactly.

5 / 5

Trigger Term Quality

Good natural trigger phrases users would actually say — "build a new app", "scaffold a project", "expand/refactor an existing one on AWS" — but common variations like "new project", "create an app", or "add a feature" are absent. Not a 3 (the terms are natural user language, not jargon) and not a 5 (synonym coverage is not comprehensive).

4 / 5

Distinctiveness Conflict Risk

Clear niche (structured discovery flow followed by code changes, AWS-focused) with an explicit exclusion list redirecting to named sibling skills — "architecture advice/recommendations without code changes (see architect-for-startups)", "agent POC (see agent-advisor)", "migrating an existing GCP/Azure/Heroku workload... (see gcp-to-aws / azure-to-aws / heroku-to-aws / llm-to-bedrock)". Conflict risk with adjacent skills is minimal.

5 / 5

Total

19

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 2 missing, 2 suspicious

Warning

Total

15

/

16

Passed

Repository

aws/agent-toolkit-for-aws

Path

plugins/aws-startup-advisor/skills/start-building-for-startups/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.