judgment-day

Trigger: judgment day, dual review, adversarial review, juzgar. Run explicit blind dual review with at most two scoped fix/re-judgment rounds.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

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.

A tightly written orchestration skill: concise, dense with operational rules, and a clear sequenced workflow with explicit validation and bounded feedback loops. The only notable issues are a couple of abstract steps (artifact-store persistence, immutable-target construction) and a dangling out-of-bundle shared reference.

Suggestions

Make the persistence and target-construction steps concrete — name the actual artifact store command/path and what "immutable target" means operationally (e.g., a tag, commit SHA, or frozen directory).

Resolve or inline the essential parts of ../_shared/review-ledger-contract.md so the skill has no dependency on a file outside its own bundle.

Dimension	Reasoning	Score
Conciseness	The body is lean and imperative throughout — hard rules, a decision table, six numbered steps, and an output contract — with no explanations of concepts Claude already knows. Every line is operational and earns its place.	5 / 5
Actionability	Guidance is largely concrete: named subagents (`jd-judge-a`, `jd-judge-b`, `jd-fix-agent`), exact verdict values (`APPROVED \| ESCALATED`), a decision-gate table, and a real reference file containing the full judge/fix prompts. It is not 5 because a few steps remain abstract — "persist it through the selected artifact store" and "build one complete immutable target" give no concrete mechanism or command.	4 / 5
Workflow Clarity	The six execution steps are clearly sequenced with explicit checkpoints (wait for both judges, ask before round-one correction, re-judgment sees only the frozen ledger plus fix delta and may record fix-caused defects), and the decision-gates table plus hard two-round budget form a feedback loop with defined escalation. It is not 4 because validation and feedback are explicit rather than implicit.	5 / 5
Progressive Disclosure	The body is a lean overview and the main reference (references/prompts-and-formats.md) is real, clearly signaled, one level deep, and well described. It is not 5 because the second reference (../_shared/review-ledger-contract.md) points outside the skill bundle and does not resolve in this environment, creating a fragile external dependency even though it is explicitly marked optional.	4 / 5
	Total	18 / 20 Passed

Description

78%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 concise, third-person description with an explicit trigger clause and a clearly bounded what. Its main weaknesses are narrow action coverage (only the top-level review loop is stated) and a few missing natural trigger variations, which leave modest overlap risk with ordinary review skills.

Suggestions

List two or three of the skill's concrete component actions in the description (e.g., launching two blind judges, merging findings into a frozen ledger, returning an APPROVED/ESCALATED verdict) to raise specificity.

Add one or two natural trigger variations users might say, such as "adversarial code review" or "second-opinion review", to broaden trigger coverage.

Sharpen the when-clause to name the target type (e.g., "for a concrete code or document target") so it is less likely to fire on generic review requests that ordinary review should handle.

Dimension	Reasoning	Score
Specificity	The description names one concrete, bounded action — "Run explicit blind dual review with at most two scoped fix/re-judgment rounds" — but coverage is narrow; the component actions of the skill (merging findings, fix rounds, terminal verdicts, escalation) are not listed. It is not 4 because it does not list several specific actions, and not 2 because the action stated is concrete and bounded rather than generic.	3 / 5
Completeness	It explicitly answers "when" via a concrete trigger clause ("Trigger: judgment day, dual review, adversarial review, juzgar") and "what" ("Run explicit blind dual review with at most two scoped fix/re-judgment rounds"). It is not 4 because the when-guidance is explicit and concrete, matching the anchor that requires concrete trigger phrases rather than an merely adequate 'Use when' clause.	5 / 5
Trigger Term Quality	"judgment day, dual review, adversarial review, juzgar" are natural phrases a user would plausibly say, including a Spanish synonym. It is not 5 because common variations such as "adversarial code review" or "second-opinion review" are missing; it is clearly above 3 since multiple relevant natural terms are present.	4 / 5
Distinctiveness Conflict Risk	"Judgment Day" blind dual review is a distinct niche with dedicated trigger terms, but "dual review" and "adversarial review" overlap ordinary code-review skills (the skill itself notes it replaces "ordinary 4R"). It is not 5 due to this residual overlap with closely related review skills.	4 / 5
	Total	16 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 1 suspicious	Warning

	Total	15 / 16 Passed

Repository: Gentleman-Programming/gentle-ai
Path: internal/assets/skills/judgment-day/SKILL.md
Commit: d96a5d4

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

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.