CtrlK
BlogDocsLog inGet started
Tessl Logo

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

74

1.05x
Quality

Does it follow best practices?

Impact

60%

1.05x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No known issues

SKILL.md
Quality
Evals
Security

Evaluation results

92%

31%

Add Dark Mode to Task Tracker

Design-before-code gate and spec writing workflow

Criteria
Without context
With context

No early implementation

100%

100%

Spec file location

0%

100%

Spec filename convention

0%

100%

Spec file is Markdown

100%

100%

Spec covers architecture

100%

100%

Spec covers components

100%

100%

Spec covers data flow

100%

100%

Spec covers error handling

100%

66%

No placeholders in spec

100%

100%

Git commit exists

0%

100%

User review prompt

50%

50%

Writing-plans handoff

33%

50%

89%

-6%

Platform Kickoff: From Idea to Design Doc

Multi-subsystem scope flagging and one-at-a-time clarifying questions

Criteria
Without context
With context

Scope flag raised

100%

100%

Questions one at a time

90%

100%

Multiple-choice questions used

100%

0%

Exactly 2-3 approaches proposed

100%

100%

Trade-offs stated per approach

100%

100%

Recommended option leads

100%

100%

Architecture section present

100%

100%

Components section present

100%

100%

Data flow section present

83%

50%

Error handling section present

83%

100%

Testing section present

66%

100%

YAGNI applied

100%

100%

Spec file location

100%

100%

0%

-15%

Portfolio Website Design — Brainstorming Session

Visual companion just-in-time offer and per-question browser/terminal routing

Criteria
Without context
With context

No upfront companion offer

100%

0%

Offer as standalone message

0%

0%

Server started with --open flag

0%

0%

Server started with --project-dir flag

0%

0%

Complete URL shared

0%

0%

Gitignore reminder

0%

0%

Server-alive check logged

0%

0%

Semantic HTML filenames

0%

0%

No filename reuse

0%

0%

HTML content fragments

0%

0%

Navigation handled in terminal

100%

0%

Waiting screen pushed on return to terminal

0%

0%

URL reminder per screen push

0%

0%

STATE_DIR/events read

0%

0%

Repository
obra/superpowers
Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

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.