CtrlK
BlogDocsLog inGet started
Tessl Logo

shape

Bridge WHAT (intent) to HOW (implementation). Use when spec is clear but approach is not. Triggers on "shape this", "how should I build", "implementation approach".

64

1.53x
Quality

54%

Does it follow best practices?

Impact

72%

1.53x

Average score across 3 eval scenarios

SecuritybySnyk

Advisory

Suggest reviewing before use

Optimize this skill with Tessl

npx tessl skill review --optimize ./hope/skills/shape/SKILL.md
SKILL.md
Quality
Evals
Security

Evaluation results

78%

57%

Design a Search and Filter System for a Content Library

AskUserQuestion formatting and batching

Criteria
Without context
With context

Conclusion-first labels

20%

80%

One-line tradeoff description

30%

90%

Plain text detail panel

0%

100%

ALL CAPS section headers

0%

100%

Short lines in detail panel

12%

87%

Dashes for bullets in panel

42%

100%

Explore further option present

0%

0%

Batched non-conflicting dimensions

20%

30%

No text walls before prompt

12%

87%

Technique names absent

100%

100%

Semantic checksum present

0%

100%

60%

4%

Shape the Offline-First Sync Architecture for a Field Operations App

Pre-mortem and second-order effects analysis

Criteria
Without context
With context

Cross-dimension tension check

100%

100%

Pre-mortem present

100%

100%

Pre-mortem: discoverability

0%

0%

Pre-mortem: accessibility

0%

0%

Pre-mortem: concurrent operations

100%

100%

Pre-mortem: maintenance burden

0%

50%

Second-order effects per dimension

100%

100%

Closed-off futures visible

100%

100%

Integration options presented

0%

0%

Exit audit not shown

100%

100%

Semantic checksum in emit

0%

0%

79%

13%

Shape the Authentication Migration for a Multi-Tenant SaaS Platform

Entry audit, classification inference, and scoped shaping

Criteria
Without context
With context

Domain type inferred, not asked

55%

100%

Reversibility inferred, not asked

100%

100%

Externalities identified

100%

100%

Externalities not shaped

100%

100%

Entry audit not shown

100%

33%

Findings as brief bullets

22%

44%

High-stakes depth applied

44%

100%

Shape document sections complete

77%

100%

First step is atomic

22%

77%

Semantic checksum present

0%

100%

No technique names in output

100%

0%

Repository
saadshahd/moo.md
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.