multi-ai-research

Comprehensive research and analysis using Claude (subagents), Gemini CLI, and Codex CLI. Multi-perspective research with cross-verification, iterative refinement, and 100% citation coverage. Use for security analysis, architecture research, code quality assessment, performance analysis, or any research requiring rigorous verification and multiple AI perspectives.

Quality

64%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/multi-ai-research/SKILL.md

Quality

Content

45%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 lays out a clear, well-sequenced multi-phase workflow but is badly overlong and redundant, with marketing claims and placeholder-driven implementation undercutting actionability. Progressive disclosure is essentially absent given the single monolithic file.

Suggestions

Collapse the three repeated analysis-category sections (When to Use, Analysis Types, Examples) into one, and remove unverifiable marketing claims like per-example 'Quality: 97/100' and the 'Quality Guarantees ✓' list.

Split detailed material into references/ files (e.g. analysis-types.md, quality-rubric.md, examples.md) and have SKILL.md point to them one level deep, instead of inlining everything.

Replace the placeholder Claude orchestration in scripts/analyze.sh (the 'In actual implementation, this would use the Task tool' block) with a real, executable invocation, or clearly document that the script is a scaffold to be driven by Claude Code.

Dimension	Reasoning	Score
Conciseness	The body is ~730 lines with heavy redundancy: analysis categories appear three times (When to Use, Analysis Types, Examples), marketing-style 'Quality Guarantees ✓' bullets and fabricated per-example quality scores (97/100, 98/100) pad the content, and phases are restated repeatedly.	2 / 5
Actionability	Install/test commands for Gemini and Codex CLIs and the analyze.sh invocation are concrete and executable, but the central orchestration is explicitly a placeholder ('In actual implementation, this would use the Task tool') and the bundled script simulates Claude research rather than implementing it.	3 / 5
Workflow Clarity	The 5/6-phase pipeline is clearly sequenced with per-phase outputs, durations, and an explicit verification-then-iterate loop, which provides good checkpoints; minor gaps come from the placeholder-driven implementation not matching the documented flow.	4 / 5
Progressive Disclosure	The skill is a monolithic ~730-line SKILL.md with no references/ files; material that belongs in separate references (full analysis-type specs, the quality rubric, worked examples) is all inlined, and the only external pointer is to the bundled scripts/analyze.sh.	2 / 5
	Total	11 / 20 Passed

Description

83%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 is well-constructed: third person, concrete systems and actions, explicit 'Use for' trigger guidance, and a clear niche. It is strong on completeness with only minor gaps in trigger-term breadth and distinctiveness at the edges.

Dimension	Reasoning	Score
Specificity	Names concrete systems (Claude subagents, Gemini CLI, Codex CLI) and several concrete actions (research, analysis, cross-verification, iterative refinement, 100% citation coverage), but the actions are somewhat high-level rather than fully enumerated.	4 / 5
Completeness	Clearly states what it does ('Comprehensive research and analysis... Multi-perspective research with cross-verification, iterative refinement, and 100% citation coverage') and explicitly answers when via 'Use for security analysis, architecture research... or any research requiring rigorous verification'.	5 / 5
Trigger Term Quality	Includes natural phrases users would say ('security analysis', 'architecture research', 'code quality assessment', 'performance analysis', 'research'), though it lacks synonyms and file-extension variants that would push it to 5.	4 / 5
Distinctiveness Conflict Risk	The multi-AI / cross-verification niche is fairly distinct from typical single-purpose skills, but the catch-all 'any research requiring rigorous verification and multiple AI perspectives' leaves some overlap with general analysis skills.	4 / 5
	Total	17 / 20 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 (736 lines); consider splitting into references/ and linking	Warning
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning

	Total	14 / 16 Passed

Repository: fernandezbaptiste/Skrillz
Path: skills/multi-ai-research/SKILL.md
Commit: 93ed392

Reviewed: 1 day 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.