literature-grounding

Use this skill when the user asks for papers, prior work, literature support, benchmark conventions, or other explicit evidence grounding from published work.

Quality

61%

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/literature-grounding/SKILL.md

Quality

Content

67%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 is a compact, decision-oriented routing policy with genuinely concrete parameters, depth tiers, and source-escalation rules, and its workflow is clearly sequenced. Its main weaknesses are redundancy — trigger conditions and depth guidance each appear twice, with 'focused' listed in only one place — and a References pointer to something that is not part of the bundle.

Suggestions

State the depth tiers once (a single quick/standard/focused/deep_report table or list) and reference it from both Quick Start and Workflow to remove the duplication and the 'focused'-only-in-one-place inconsistency.

Consolidate the trigger conditions from Quick Start, Workflow step 1, and 'Method-critical defaults' into one section; the current triple statement spends tokens without adding information.

Clarify the References entry: mark 'catalysis-prior-art-and-benchmarking' explicitly as a separate skill (or remove/replace it with a real bundle path), since no references/ directory exists in this skill.

Dimension	Reasoning	Score
Conciseness	The body is operational policy rather than concept tutorial, but the same guidance repeats: trigger conditions appear in Quick Start, Workflow step 1, and 'Method-critical defaults', and depth routing is stated twice with 'focused' appearing only in the Workflow copy (inconsistent with Quick Start). This matches anchor 3's 'mostly efficient but could be tightened'; not 2 because nothing explains what Claude already knows, not 4 because the redundancy is more than minor.	3 / 5
Actionability	Concrete, executable guidance: names the tool run_literature_research, its query/topic parameters, enumerated depth values (quick/standard/focused/deep_report), and explicit source escalation (web-first, then OpenAlex / Semantic Scholar when paper metadata is needed). As an instruction-only skill the absence of code is not penalized; minor gaps (no example invocation, 'focused' missing from Quick Start) keep it at anchor 4 rather than 5.	4 / 5
Workflow Clarity	A clearly sequenced 4-step workflow (decide need, choose smallest depth, route sources, keep pack clean) with an escalation checkpoint ('when the web pass is too weak') and explicit do-not rules. Matches anchor 4; not 5 because checkpoints are implicit conditions rather than explicit verify/validate steps, though no destructive or batch operations exist so the validation cap does not apply.	4 / 5
Progressive Disclosure	The ~42-line body is well organized into Overview/Quick Start/Workflow/defaults/Output Contract sections, which for a short skill with no bundle files would justify a 5. However the References section points to 'catalysis-prior-art-and-benchmarking', which is not a file in this bundle and is not clearly signaled as a separate skill versus a bundle reference — a minor navigation gap matching anchor 4.	4 / 5
	Total	15 / 20 Passed

Description

56%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 has strong, natural trigger coverage and a distinctive niche, but it reads as only a 'when' clause: it never states what the skill actually does (decide literature depth, route run_literature_research), leaving the what half of completeness weak and its actions generic. A single leading capability sentence would lift both completeness and specificity.

Suggestions

Prepend a concrete 'what' clause in third person, e.g. 'Decides when a request needs published-work grounding and selects the literature-research depth and sources. Use when the user asks for papers, prior work...'

Add one or two concrete capability verbs (e.g. 'compiles a citation pack', 'routes queries to OpenAlex / Semantic Scholar') so the description states actions, not just triggers.

Add commonly used synonyms such as 'citations', 'related work', or 'academic papers' to round out trigger coverage toward the top anchor.

Dimension	Reasoning	Score
Specificity	The description is a pure trigger clause ('papers, prior work, literature support, benchmark conventions... evidence grounding from published work') that names the domain but states no concrete actions. It is above anchor 1 because the domain is specific, but below anchor 3 because no 1-2 concrete capabilities are listed.	2 / 5
Completeness	The 'when' is explicit and strong ('Use this skill when the user asks for...') but the 'what' is only weakly implied by 'evidence grounding from published work' — the skill's actual behavior (depth selection, routing run_literature_research) never appears. This mirrors anchor 3's missing-half pattern; not 2 because a what is at least implied, not 4 because that what is thin.	3 / 5
Trigger Term Quality	Natural user phrases with synonym coverage ('papers', 'prior work', 'literature support', 'benchmark conventions', 'published work'), matching anchor 4's 'good keyword coverage; a few natural terms missing' — variants like 'citations', 'related work', and 'academic papers' are absent, so not 5.	4 / 5
Distinctiveness Conflict Risk	The paper-grounding niche ('papers, prior work, literature support, benchmark conventions') is mostly distinct from generic skills, matching anchor 4's 'minor overlap risk' — terms like 'papers' and 'literature' could also trigger a general research/deep-research skill, so not 5.	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: fernandezbaptiste/CatMaster
Path: skills/literature-grounding/SKILL.md
Commit: c87fc76

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.