novelty-check

Verify research idea novelty against recent literature. Use when user says "查新", "novelty check", "有没有人做过", "check novelty", or wants to verify a research idea is novel before implementing.

Quality

—

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Advisory

Suggest reviewing before use

Quality

Content

77%

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

A well-structured, actionable multi-phase workflow with strong validation feedback loops. It loses points for verbosity in a few sections and for relying on external reference files that are not present in the bundle and are reached through an indirection layer.

Suggestions

Tighten the Phase C dossier paragraph and the report template to reduce token overhead, keeping only the fields Claude cannot infer.

Inline or vendor the critical rules from shared-references/*.md (especially the citation-discipline pre-search verification steps) so the skill is self-contained when those files are absent, or confirm those files ship with the bundle.

Collapse the integration-contract resolver indirection into direct file references to keep navigation strictly one level deep.

Dimension	Reasoning	Score
Conciseness	The body is mostly lean and avoids explaining concepts Claude already knows, but the Phase C dossier instructions, the six-bullet "Important Rules" list, and the fairly elaborate report template could be tightened, placing it at the mostly-efficient anchor 2 rather than the every-token-earns-its-place anchor 3.	2 / 3
Actionability	Provides a copy-paste-ready MCP invocation block with a concrete config, specific search parameters (year filters 2024-2026, "at least 3 query formulations", named venues), and a concrete deliverable template, matching the executable/copy-paste-ready anchor 3 rather than the pseudocode anchor 2.	3 / 3
Workflow Clarity	Phases A-D are clearly sequenced and include explicit verification checkpoints (pre-search verification via verify_papers.py with a degraded-output fallback policy D1, and mandatory review tracing), matching the clear-sequence-with-validation anchor 3 rather than the missing-checkpoints anchor 2.	3 / 3
Progressive Disclosure	References are well-signaled with markdown links and section numbers and are mostly one level deep, but no bundle files exist to verify the referenced paths (../shared-references/*.md, verify_papers.py, save_trace.sh), and integration-contract.md is used as a resolver chain, introducing indirection that fits the could-be-better-organized anchor 2 rather than the clean one-level anchor 3.	2 / 3
	Total	10 / 12 Passed

Description

90%

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 strong description with explicit what/when structure, natural multilingual triggers, and a distinct niche. Its only weakness is that it states one overarching action rather than listing several concrete capabilities.

Dimension	Reasoning	Score
Specificity	Names the domain and a single concrete action ("Verify research idea novelty against recent literature") but does not enumerate multiple specific actions like literature search, cross-model verification, or report generation, so it falls short of the multi-action anchor 3.	2 / 3
Completeness	Explicitly answers both what ("Verify research idea novelty against recent literature") and when via a clear "Use when..." clause with explicit triggers, matching the anchor 3 example; not 2 because the when-clause is explicit, not merely implied.	3 / 3
Trigger Term Quality	Provides several natural trigger phrases a user would actually say across two languages ("查新", "novelty check", "有没有人做过", "check novelty"), matching the good-coverage anchor 3 rather than the partial-coverage anchor 2.	3 / 3
Distinctiveness Conflict Risk	The research-novelty niche is narrow and the triggers are highly specific, making it unlikely to fire for unrelated skills, matching the clear-niche anchor 3 rather than the overlapping anchor 2.	3 / 3
	Total	11 / 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 — 13 / 16 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
relative_links	Relative link issues: 2 suspicious	Warning

	Total	13 / 16 Passed

Repository: wanshuiyin/Auto-claude-code-research-in-sleep
Commit: fe5963c

Reviewed: about 16 hours 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.