CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

elicit

Draw out what the user holds but can't say. Use when the user knows something they cannot yet express.

Quality

76%

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/elicit/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

78%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.

This is a tight, instruction-only skill that specifies a concrete interactive method with an explicit feedback loop and stopping conditions, and it correctly keeps everything in one short file. Its main weaknesses are the absence of a worked example, a convoluted transition sentence around AskUserQuestion, and slight redundancy in the direct-question guidance.

Suggestions

Add one compact example of an offer and its reaction (e.g. a candidate and a 'closer but' reply showing how the record is updated) to make the method copy-paste concrete.

Rewrite the transition sentence in plain terms — e.g. 'When the record settles everything reachable sources can settle, ask the user directly (AskUserQuestion) what it still leaves open' — to remove the checkpoint ambiguity.

Merge paragraph 3 with the routing paragraph so the direct-question rule is stated once, trimming the redundancy.

DimensionReasoningScore

Conciseness

The body is lean — roughly 250 words for a complete elicitation method, assuming Claude's competence with no concept explanations — but there is minor redundancy: paragraph 3 ("If a reaction shows the user would answer a direct question, ask it instead") restates the first clause of the final routing paragraph. This is anchor 4 (efficient with minor instances that could be trimmed), not anchor 5.

4 / 5

Actionability

Guidance is concrete for an instruction-only skill: "offer one concrete thing the user can react to — a candidate, a contrast, a small set laid side by side", reaction types ("a yes, a no, a 'closer but'"), a named tool (AskUserQuestion), and a defined deliverable (hand back the record in confirmed words). It falls short of anchor 5 because there is no worked example of an offer or a short exchange, which would cover the most common case.

4 / 5

Workflow Clarity

The sequence is clear with a genuine feedback loop (offer → read reaction → update record → adjust) and explicit convergence criteria ("until a new offer stops changing the record, or the user says the record matches what they hold"). It is anchor 4 rather than 5 because the transition checkpoint — "Ask what the record leaves open, with AskUserQuestion, once it holds what sources in reach settle" — is convoluted and leaves the handoff moment implicitly defined.

4 / 5

Progressive Disclosure

The skill is under 50 lines with no content that belongs in separate files, so a single well-organized SKILL.md is the correct structure per the simple-skill guideline. Cross-references to the **clarify** and **explain** skills are explicit and clearly signaled, and there are no nested or buried references.

5 / 5

Total

17

/

20

Passed

Description

62%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 is concise, has an explicit 'Use when…' trigger, and carves out a distinct niche, but it states its capability abstractly and offers only one trigger phrasing. It reads more as a mission statement than a capability inventory, which limits specificity and trigger coverage.

Suggestions

Name the concrete mechanisms in the description itself, e.g. 'Elicit tacit knowledge by offering concrete candidates to react to and keeping a running record in the user's own words', so the 'what' lists actions rather than one abstract outcome.

Add one or two natural trigger variations users would actually say, such as "can't put it into words" or "I know it when I see it", to broaden keyword coverage.

Add a short contrast clause (e.g. 'not for answering direct questions or disambiguating wording') to reduce overlap with clarify/explain-style skills at the description level.

DimensionReasoningScore

Specificity

"Draw out what the user holds but can't say" states one capability but names no concrete mechanisms — the actual actions (offering candidates/contrasts, keeping a running record) appear only in the body. This matches anchor 3 ("names domain and 1-2 concrete actions") more than anchor 2, because the phrase precisely defines the function rather than being generic filler, but it is not the multiple concrete actions of anchor 4.

3 / 5

Completeness

Both parts are present: the "what" ("Draw out what the user holds but can't say") and an explicit "Use when…" clause for the "when". It falls short of anchor 5 because the "what" is a single abstract capability rather than concrete trigger phrases and listed capabilities — the "when" could also cover more situations.

4 / 5

Trigger Term Quality

The trigger phrase "Use when the user knows something they cannot express" is a natural way a user might describe their state, but it is the only variation given. Missing common synonyms users actually say ("can't put it into words", "I know it when I see it", "can't articulate it"), matching anchor 3 ("some relevant keywords but missing common variations or synonyms").

3 / 5

Distinctiveness Conflict Risk

The niche (eliciting knowledge the user holds but cannot express) is distinct and the trigger is specific, giving minimal conflict risk with unrelated skills. It scores 4 rather than 5 because the description alone does not disambiguate from closely related conversational skills (e.g., clarify-style questioning); that disambiguation lives in the body.

4 / 5

Total

14

/

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/elicit/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.