best-of-n

Generate a small set of independent candidate solutions in worktrees, judge them against one explicit rubric, and apply the winner only after PASS verification.

Quality

64%

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

Fix and improve this skill with Tessl

tessl review fix ./crates/tui/assets/skills/best-of-n/SKILL.md

Quality

Content

77%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 content is a well-sequenced, lean, actionable workflow with strong verification checkpoints, but its single external file reference is unverifiable and not organized under a references/ directory.

Suggestions

Verify the referenced 'operate_best_of_n.workflow.js' exists and place it under a references/ directory (or remove the reference if it is unavailable).

Turn the JSON action example into a complete, copy-paste-ready snippet (including the collect/wait step) so the common path is fully executable.

Consider moving the long tournament-setup and search-concurrency details into a separate reference file to keep the overview shorter.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence without explaining basic concepts, with only a few explanatory caveats that could be trimmed.	4 / 5
Actionability	It provides a concrete JSON action contract and explicit builder field guidance, though the example is a partial action block rather than a fully copy-paste-ready end-to-end script.	4 / 5
Workflow Clarity	Four phases are clearly sequenced with explicit validation checkpoints and feedback loops ('Integrate Only After PASS', runtime-run hard gates, independent replay, stop-early rule) appropriate for a batch/verification workflow.	5 / 5
Progressive Disclosure	Structure and section headers are good, but the one external reference ('operate_best_of_n.workflow.js') is neither in a references directory nor present in the workspace, leaving a dangling, unsignaled pointer.	3 / 5
	Total	16 / 20 Passed

Description

51%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 concretely states the skill's three actions and occupies a distinct niche, but omits any explicit 'when to use' trigger clause and leans on technical jargon rather than natural user keywords.

Suggestions

Append a 'Use when...' clause with concrete trigger phrases such as 'when a task has several plausible solutions and comparison is worth the extra model work'.

Add natural user-facing keywords (e.g., 'compare candidates', 'pick the best approach') so the description matches language users actually say.

Optionally surface the default N=3 and the 2-4 quick-comparison range in the description to improve coverage.

Dimension	Reasoning	Score
Specificity	Three concrete actions are named explicitly ('Generate a small set of independent candidate solutions in worktrees', 'judge them against one explicit rubric', 'apply the winner only after PASS verification'), with only minor coverage gaps.	4 / 5
Completeness	It states a clear 'what' but provides no 'Use when...' trigger clause, which caps completeness at 3 per the rubric guidance.	3 / 5
Trigger Term Quality	The description relies on jargon ('worktrees', 'rubric', 'PASS verification') and lacks natural user phrases like 'compare options' or 'pick the best approach', which only appear in the body.	2 / 5
Distinctiveness Conflict Risk	The ensemble/tournament niche is clearly distinguishable and unlikely to misfire, with only minor overlap risk against general review or comparison skills.	4 / 5
	Total	13 / 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
metadata_version	'metadata.version' is missing	Warning

	Total	15 / 16 Passed

Repository: Hmbown/Codewhale
Path: crates/tui/assets/skills/best-of-n/SKILL.md
Commit: 21282f1

Reviewed: 6 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.