skill-debate

Structured multi-provider AI debates between Claude and available advisors — use for critical decisions

Quality

56%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/skill-debate/SKILL.md

Quality

Content

62%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable and well-sequenced with strong validation feedback loops, but it is severely verbose — duplicating whole sections and over-emphasizing compliance — and does not use progressive disclosure to move detail into bundled files. Token efficiency is the main weakness.

Suggestions

Remove the duplicated sections: the activation banner, quality-gates table, cost tracking, and export blocks each appear twice — keep one canonical copy and cross-reference it.

Move detailed reference material (flag tables, quality-gate metrics, cost estimates, export options) into bundled reference files under references/ and link to them from a lean overview, reducing the SKILL.md body substantially.

Cut the compliance shouting and explanatory asides (e.g. the rationale for using Sonnet over Opus) to assume Claude's competence and reclaim context-window budget.

Dimension	Reasoning	Score
Conciseness	The ~688-line body is padded with duplicated sections (the activation banner, quality-gates table, cost tracking, and export blocks each appear twice), heavy "MANDATORY/PROHIBITED" shouting, and explanatory asides Claude already knows (e.g. why Sonnet differs from Opus), matching the verbose/padded anchor.	1 / 3
Actionability	It provides concrete, executable guidance throughout — real script paths (orchestrate.sh spawn, check-providers.sh, build-fleet.sh), a full flag table, heredoc templates, a quality-gate bash function, and a specific Agent subagent call — reaching the copy-paste-ready bar.	3 / 3
Workflow Clarity	Steps 1–7.5 are clearly sequenced with explicit validation checkpoints (provider availability check, quality-gate thresholds with re-prompt feedback loop) and a closing quality checklist, matching the clear-sequence-with-feedback-loops anchor.	3 / 3
Progressive Disclosure	The skill is a single monolithic file with no bundled reference files (references/scripts/assets are absent), and the paths it does cite point to an external plugin rather than this skill's bundle; section structure exists but large reference-style blocks (flag tables, quality gates, examples) are inlined rather than split out.	2 / 3
	Total	9 / 12 Passed

Description

50%

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 competent and correctly pairs a what-statement with a use-when clause, but the trigger language is generic and the capability list is thin. It is distinguishable yet not sharply targeted to natural user phrasing.

Suggestions

Replace the generic "use for critical decisions" with concrete trigger phrases users actually say, e.g. "Use when the user wants a debate, a second opinion from other models, or to compare perspectives on a tough decision."

Add one or two more concrete actions to the what-clause, e.g. "dispatches external advisors, runs structured rounds, and synthesizes their positions".

Tie the when-clause to debate-specific scenarios (multi-model review, adversarial analysis, trade-off comparison) to reduce overlap with generic decision skills.

Dimension	Reasoning	Score
Specificity	"Structured multi-provider AI debates between Claude and available advisors" names the domain and a core action (debates) but does not enumerate multiple concrete actions (e.g. dispatch, synthesize, score responses), so it is not comprehensive enough for a 3.	2 / 3
Completeness	It states both what (multi-provider AI debates) and when ("use for critical decisions"), but the when-clause is generic and not tied to the skill's actual debate scenarios, keeping it below the explicit-specific-trigger bar for a 3.	2 / 3
Trigger Term Quality	It includes relevant terms like "debates" and "critical decisions", but misses common natural variations users would say ("second opinion", "compare options", "get multiple perspectives", "review with other models"), so coverage is partial.	2 / 3
Distinctiveness Conflict Risk	The "multi-provider AI debates" niche is fairly distinct, but the generic "critical decisions" trigger could overlap with general decision-making or review skills, so conflict risk is not fully eliminated.	2 / 3
	Total	8 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (688 lines); consider splitting into references/ and linking	Warning
referenced_paths_exist	Referenced path issues: 1 missing, 1 deeper-than-1-level	Warning

	Total	14 / 16 Passed

Repository: nyldn/claude-octopus
Path: .claude/skills/skill-debate/SKILL.md
Commit: 9f39f0c

Reviewed: 2 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.