paper-illustration-image2

Generate publication-quality academic illustrations through a local Codex app-server bridge that uses Codex native image generation. This is a separate experimental alternative to `paper-illustration`, intended for Claude Code users who want a GPT-image-style renderer without modifying the original skill.

Quality

60%

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./skills/paper-illustration-image2/SKILL.md

Quality

Discovery

57%

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 does a good job distinguishing itself from a related skill ('paper-illustration') and identifies its niche (Codex-based image generation for academic illustrations). However, it lacks specific concrete actions beyond 'generate illustrations', misses natural user trigger terms, and has no explicit 'Use when...' clause to guide skill selection.

Suggestions

Add an explicit 'Use when...' clause, e.g., 'Use when the user asks for academic figures, scientific diagrams, or paper illustrations and wants to use the Codex image generation pipeline.'

Include natural trigger terms users would say, such as 'figure', 'diagram', 'scientific illustration', 'research figure', 'paper graphic', '.png', '.svg'.

List more specific concrete actions beyond just 'generate', e.g., 'Creates diagrams, schematics, conceptual figures, and data visualizations for academic papers.'

Dimension	Reasoning	Score
Specificity	It names the domain ('publication-quality academic illustrations') and mentions the mechanism ('local Codex app-server bridge', 'Codex native image generation'), but doesn't list multiple concrete actions beyond 'generate illustrations'. The description focuses more on architecture than specific capabilities.	2 / 3
Completeness	The 'what' is partially addressed (generate academic illustrations via Codex bridge), and there's an implicit 'when' (Claude Code users wanting GPT-image-style rendering), but there is no explicit 'Use when...' clause with clear trigger guidance. The description focuses more on differentiating from 'paper-illustration' than on when to use it.	2 / 3
Trigger Term Quality	Includes some relevant terms like 'academic illustrations', 'publication-quality', 'GPT-image-style', and 'Codex', but misses many natural user terms like 'figure', 'diagram', 'scientific figure', 'paper figure', 'research illustration'. The technical jargon ('app-server bridge') is not what users would naturally say.	2 / 3
Distinctiveness Conflict Risk	The description explicitly differentiates itself from the 'paper-illustration' skill, specifying it as an 'experimental alternative' using a distinct mechanism (Codex app-server bridge). This makes it clearly distinguishable and reduces conflict risk with the related skill.	3 / 3
	Total	9 / 12 Passed

Implementation

62%

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

The skill provides a thorough, well-structured workflow with strong actionability and clear validation checkpoints, making it effective for its intended purpose. However, it is significantly over-verbose: the large ASCII diagram duplicates the step-by-step instructions, the style guide explains conventions Claude already knows, and the helper resolution script is excessively detailed inline. The content would benefit greatly from splitting the style guide and helper resolution into separate referenced files.

Suggestions

Move the CVPR/NeurIPS style guide into a separate referenced file (e.g., STYLE_GUIDE.md) and keep only a 2-3 line summary inline, cutting ~60 lines of content Claude largely already knows.

Remove or drastically shorten the large ASCII workflow diagram — the numbered steps already convey the same information more concisely.

Move the IMAGE2_HELPER resolution script into the helper script itself or a setup document rather than inlining 25+ lines of bash fallback logic in the skill body.

Remove the Chinese text line and redundant restatements of rules (e.g., arrow standards appear in both the style guide and the key rules section).

Dimension	Reasoning	Score
Conciseness	The skill is extremely verbose at ~300+ lines. The ASCII art workflow diagram is massive and redundant with the step-by-step instructions that follow. The style guide section explains basic academic figure conventions at length (what CVPR style means, what to avoid) that Claude already knows. The IMAGE2_HELPER resolution script is overly detailed with multiple fallback layers. Mixed Chinese text ('目标：既不保守也不花哨，找到平衡点') adds noise. Many rules are restated multiple times across sections.	1 / 3
Actionability	The skill provides fully executable bash commands for preflight, finalize, and verify steps. MCP tool calls are specified with exact parameter names. The LaTeX snippet is copy-paste ready. Refinement feedback examples are concrete and specific. The helper resolution script, while verbose, is executable.	3 / 3
Workflow Clarity	The multi-step workflow is clearly sequenced (Steps 0-7) with explicit validation checkpoints: preflight check with ok=true gate, score-based acceptance threshold (≥9), iterative refinement loop with concrete feedback, and a final verify step before claiming success. The repair path for skipped artifacts is a nice addition. Missing feedback loops would cap at 2, but this skill has them.	3 / 3
Progressive Disclosure	The skill references an integration contract and helper scripts in external paths, but the body itself is monolithic — the entire CVPR style guide, the full helper resolution script, the complete workflow, constants, scope table, key rules, repair path, and output structure are all inline. The style guide and helper resolution logic could easily be split into separate referenced files. No bundle files are provided to offload content to.	2 / 3
	Total	9 / 12 Passed

Validation

81%

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 — 9 / 11 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	9 / 11 Passed

Repository: wanshuiyin/Auto-claude-code-research-in-sleep
Commit: 66b974e

Reviewed: 3 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.