CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

draft

Build the thing the user can't specify, choices surfaced. Use when the user needs something to react to before they can choose.

Quality

78%

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 ./hope/skills/draft/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

88%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 body is an exceptionally lean, instruction-dense workflow: it defines the artifact policy (smallest complete version, fork handling), the surfacing contract, an explicit interaction sequence with a wait-for-reaction feedback loop, and conditional routing to sibling skills. Its only gaps are the absence of a worked example of a surfaced choice and the lack of section headers to organize the four short paragraphs.

DimensionReasoningScore

Conciseness

The ~230-word body is lean and directive throughout — "Build the smallest complete version — real enough to react to, small enough to throw away" — with zero explanation of concepts Claude already knows and every sentence instructing. The apparent restatement "Build, surface, hand back, once; then wait for the reaction" adds new sequencing information (single pass, then wait), so it is not trimmable padding; this matches the anchor-5 example of every token earning its place.

5 / 5

Actionability

Guidance is concrete for an instruction-only skill: it names the tool ("Ask the surfaced choices first, with AskUserQuestion"), the ordering rule ("the one you are least sure of first"), the delivery format ("its content in the reply, and where it lives"), and explicit surfacing requirements ("state the choice, what you picked, and what would make you pick otherwise"). It falls short of 5 because there is no worked example of a surfaced choice or hand-back, which the top anchor expects to cover common cases.

4 / 5

Workflow Clarity

The multi-step process is explicitly sequenced — build, surface, ask (least-sure first), hand back, wait — with ordering markers ("first", "Then", "once; then wait") and an explicit feedback loop: "then wait for the reaction. Every surfaced choice stays open until the user reacts to it." The pre-build branching paragraph gives clear conditional routing ("→ use **clarify** skill" etc.), matching the top anchor's clear sequence with explicit feedback loops; no validation checkpoints are needed since nothing destructive or batch is performed.

5 / 5

Progressive Disclosure

The body is appropriately sized for a single-file skill (no references/, scripts/, or assets/ exist, and nothing here needs to be split out), and the cross-skill pointers are clearly signaled with bolded names and arrow conditions. It scores 4 rather than 5 because the simple-skill exception requires "well-organized sections," and the content is four unheaded prose paragraphs — a minor organization gap (e.g., no headers for Build / Surface / Hand back / Routing).

4 / 5

Total

18

/

20

Passed

Description

57%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.

The description has a clear structure with an explicit "Use when" clause and a distinctive, memorable trigger condition. Its weakness is specificity: the capability is stated as "Build the thing the user can't specify," which names no concrete deliverable or domain, and trigger coverage is a single narrow phrase without common synonyms like draft, prototype, or mockup.

Suggestions

Name the concrete deliverable in the what-clause — e.g., "Build a draft, mockup, or prototype of the thing the user can't specify" — so the artifact type is explicit rather than "the thing".

Add natural trigger synonyms to the when-clause: "Use when the user needs something to react to — a first draft, mockup, or prototype — before they can choose," covering the phrases users actually say.

Spell out the second capability as an action ("surfaces open choices with what would make you pick otherwise") instead of the compressed fragment "choices surfaced" to lift both specificity and completeness.

DimensionReasoningScore

Specificity

"Build the thing the user can't specify" names one real action verb but its object is maximally generic — no domain or artifact type is stated — and "choices surfaced" is a compressed fragment rather than a concrete capability. It is not a 1 because "Build" is an explicit action, but it falls short of anchor 3, which expects 1-2 concrete, domain-anchored actions.

2 / 5

Completeness

Both parts are present: a what ("Build the thing the user can't specify, choices surfaced") and an explicit when ("Use when the user needs something to react to before they can choose"). It is capped below 5 because the what is vague about the deliverable and the when lacks concrete trigger phrases that anchor 5 requires; it is above 3 because an explicit "Use when" clause exists.

4 / 5

Trigger Term Quality

The when-clause "Use when the user needs something to react to before they can choose" contains one genuinely natural phrase users do say, so it clears anchor 2. But it misses all common variations and synonyms — draft, prototype, mockup, sketch, first pass, "I don't know what I want" — so it does not reach anchor 4's "good keyword coverage".

3 / 5

Distinctiveness Conflict Risk

The trigger condition "needs something to react to before they can choose" carves a distinctive niche (reactive drafting under uncertainty) that most build skills would not claim. Overlap risk is minor rather than negligible, since "Build the thing the user can't specify" is broad enough to collide with general creation/drafting skills when a user is vague, keeping it below anchor 5.

4 / 5

Total

13

/

20

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

saadshahd/moo.md

Path

hope/skills/draft/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.