CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

jev-gate

Review a diff with Jev, score a UI Then, or give Jev a browser goal. Trigger: "review this", "open this page", "click through with Jev", UI scenario Then.

Quality

73%

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 ./ywai/plugins/jev-gate/skills/jev-gate/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

75%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 a strong example of plugin-skill content: dense with non-obvious, measured facts about a proprietary tool, highly actionable tool names and snippets, and explicit rules for interpreting ambiguous outcomes. Its only weaknesses are minor — prose-only coverage of a few tools and slight redundancy between reporting rules.

DimensionReasoningScore

Conciseness

The body never explains concepts Claude already knows; every section delivers tool-specific, non-inferable facts ('It catches 6 of 8 fixtures', 'Two dimensions in that table never open a finding on their own'). A few sentences could still be trimmed (rules 4 and 5 overlap slightly), so it sits at anchor 4 ('efficient; minor instances of over-explanation') rather than anchor 5.

4 / 5

Actionability

All six tools are named exactly, two executable TS snippets are provided, and behaviors are specified concretely ('at most 200 segments', runId, root, the literal 'Search incomplete' marker). Not every tool gets a code example — jev_route, jev_report, and jev_find are described in prose only — which keeps it at anchor 4 rather than 5.

4 / 5

Workflow Clarity

Each capability has a clear use → interpret → next-step flow with explicit failure handling ('A missing key or a failed call is not PASS', 'Narrow it with root and run it again'), and no destructive or batch operations exist that would require a validation cap. It lacks the full checklists and structured feedback loops of anchor 5, matching anchor 4 ('clear sequence with most checkpoints present').

4 / 5

Progressive Disclosure

No bundle files exist and none are needed; the body is cleanly sectioned with descriptive headers and all decision-critical content (rules, known weaknesses, gate behavior) is appropriately inline. At ~122 lines it exceeds the under-50-line simple-skill exception for a score of 5, so anchor 4 ('good structure; most content appropriately placed') is the best fit.

4 / 5

Total

16

/

20

Passed

Description

58%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 the right structure — concrete actions plus an explicit trigger clause — but under-represents the skill's actual capabilities (routing, semantic search, reporting) and leans on trigger phrases so generic they invite collisions with other review/browser skills. Its weakest dimensions are trigger term coverage and distinctiveness.

Suggestions

Enumerate the main capabilities in the 'what' clause (e.g. routing tasks to agents, semantic code search, review reporting) so coverage matches the tools the skill actually documents.

Add natural trigger synonyms users would say, such as 'code review this diff', 'score this scenario step', or 'find where this is implemented', to broaden trigger term coverage.

Reduce conflict risk by pairing the generic phrase "review this" with a Jev-specific qualifier (e.g. "review this with Jev") and adding more distinctive trigger phrases.

DimensionReasoningScore

Specificity

The description names three concrete actions ('Review a diff with Jev, score a UI Then, or give Jev a browser goal') but coverage is not comprehensive: routing (jev_route), semantic code search (jev_find), and the review report/gate are absent, and 'give Jev a browser goal' is abstract. This matches anchor 3 ('names domain and 1-2 concrete actions, but not comprehensive') better than anchor 4, whose gaps would have to be minor.

3 / 5

Completeness

Both a 'what' (three capabilities) and an explicit 'when' (Trigger: "review this", "open this page"...) are present, so it clears the anchor-3 cap. It falls short of anchor 5 because the 'what' is compressed and jargon-laden ('score a UI Then') rather than clearly spelled out, matching anchor 4's 'when could be more explicit or specific' tier.

4 / 5

Trigger Term Quality

Phrases like "review this" and "open this page" are natural user language, but common variations are missing (e.g. 'code review this diff', 'find where X is implemented', 'route this to an agent') and 'UI scenario Then' is Gherkin jargon users would rarely say. Anchor 3 ('some relevant keywords but missing common variations or synonyms') is the closest fit; it lacks the broad natural-term coverage of anchor 4.

3 / 5

Distinctiveness Conflict Risk

'Jev' and "click through with Jev" are distinctive, but "review this" and "open this page" are extremely common phrases that would also trigger generic code-review and browser skills. Anchor 3 ('somewhat specific but could still overlap with similar skills') fits; it is not anchor 2 (broad domain claim) nor anchor 4 (only minor overlap with closely related skills).

3 / 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

YoizenSA/dev-ai-workflow

Path

ywai/plugins/jev-gate/skills/jev-gate/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.