CtrlK
BlogDocsLog inGet started
Tessl Logo

aiconfig-online-evals

DEPRECATED redirect — this skill was renamed to online-evals. Do not use this skill; invoke online-evals instead. Kept only so old references to aiconfig-online-evals still point users to the new name.

66

Quality

83%

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

SKILL.md
Quality
Evals
Security

Quality

Content

100%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 exemplary redirect stub: maximally lean, unambiguous about the single required action, and complete for both arrival paths without any filler. All four dimensions land on the top anchors because the stub does exactly one thing and states it without waste.

DimensionReasoningScore

Conciseness

The body is six lines with zero padding: 'Do not follow any instructions here and do not perform the task from this skill' and 'Invoke the online-evals skill instead' each do necessary work, and it never explains what deprecation or renaming is. It matches anchor 5 — lean, assumes Claude's competence, every token earns its place — rather than anchor 4, which requires at least minor trimmable over-explanation.

5 / 5

Actionability

The guidance is fully executable for an instruction-only skill: 'Invoke the **online-evals** skill instead' and the concrete remediation 'update it from aiconfig-online-evals to online-evals' cover both arrival paths (direct invocation and stale references). Per the scoring notes, absence of code is not penalized when guidance is actionable; both common cases are explicitly covered, matching anchor 5.

5 / 5

Workflow Clarity

This is a simple single-purpose skill whose one action — redirect to online-evals and optionally update stale references — is unambiguous and sequenced (stop, invoke replacement, fix the reference), so the simple-skill exception applies. No destructive or batch operations are involved, so no validation cap applies; it matches anchor 5's clarity rather than anchor 4, which would require a minor gap in the sequence.

5 / 5

Progressive Disclosure

The skill is under 50 lines, needs no external references (verified: no references/, scripts/, or assets/ files exist and no path in the body points to a nonexistent file), and its only pointer — the online-evals skill — is clearly signaled and one level deep. Per the under-50-line guideline this matches anchor 5 with a short, well-organized body.

5 / 5

Total

20

/

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.

A well-executed deprecation stub: unambiguous about its retired status, the replacement name, and why it still exists, with minimal conflict risk since it explicitly refuses invocation. As a capability description, however, it lacks any 'Use when...' trigger guidance and carries no natural trigger vocabulary beyond the two skill names, which limits discovery and completeness.

Suggestions

Add an explicit use-when clause that redirects trigger intent, e.g. 'Use the online-evals skill when running online evaluations or model eval tasks' — this would lift completeness above the anchor-3 cap.

Include the domain keywords a user would naturally say (e.g. 'evals', 'online evaluations', 'model evaluation') so search and skill matching still route to the right skill even when only the old name is remembered.

Briefly state what the skill domain covers (even one clause, e.g. 'this skill covered online evaluation workflows') so the redirect is informative rather than purely navigational.

DimensionReasoningScore

Specificity

The description concretely states its function ("DEPRECATED redirect", "renamed to online-evals", "Kept only so old references... still point users to the new name") but describes no capability actions of the skill itself. It fits anchor 3 — a couple of concrete, specific actions around the redirect, but no capability coverage — rather than anchor 2 (its actions are specific, not generic) or anchor 4 (no 'several specific actions' of actual functionality).

3 / 5

Completeness

The 'what' is explicit — this is a deprecated redirect stub — but there is no 'Use when...' clause or equivalent trigger guidance; the 'when' is only weakly implied by 'Kept only so old references to aiconfig-online-evals still point users to the new name'. Per the judging guideline, a missing 'Use when...' clause caps completeness at anchor 3 (clear what, when missing or weakly implied).

3 / 5

Trigger Term Quality

It includes both skill names ("online-evals", "aiconfig-online-evals") plus "DEPRECATED" and "renamed", so the exact old-name trigger is covered, but it offers no natural domain vocabulary or synonyms users would say beyond the two names. This matches anchor 3 ('some relevant keywords but missing common variations or synonyms') rather than anchor 4's good coverage.

3 / 5

Distinctiveness Conflict Risk

"Do not use this skill; invoke online-evals instead" makes wrong-skill invocation essentially impossible: the description actively disclaims use and hands off to the exact replacement, and it only matches the exact old name. This matches anchor 5 (clear niche with distinct triggers; minimal conflict risk) — nothing in it overlaps with other skills' triggers.

5 / 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
launchdarkly/ai-tooling
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.