Name: research-lookup
Rating: 67.2 (1 reviews)
Author: K-Dense-AI

research-lookup

Compile current scholarly evidence for a scientific manuscript or research brief. Use when the user explicitly asks to gather literature, references, background evidence, competing findings, or a manuscript research packet. Uses Parallel Search by default, Parallel Extract for source verification, Parallel Research for explicitly deep/exhaustive work, optional explicit Parallel Chat, and optional Perplexity only when requested or allowed as a failure fallback.

Quality

81%

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

The canonical home for this skill is research-lookup in K-Dense-AI/scientific-agent-skills

Quality

Content

75%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 highly actionable with verified, copy-paste-ready commands and a well-sequenced workflow including verification and failure-handling feedback loops. Its main weaknesses are redundant routing restatements that hurt conciseness and a monolithic structure that keeps reference-grade detail inline.

Suggestions

Consolidate the backend routing rules so they appear once (table or bullets) and remove the repeated 'explicit only' reminders in each backend section.

Move the 10 reference-quality rules and the full CLI flag catalog into a separate REFERENCES.md linked from the body to improve progressive disclosure and reduce SKILL.md length.

Add an explicit 'if Extract verification fails or the reference shortfall persists -> refine the query/date range/domains and re-run' loop directly inside step 2/3 of the manuscript workflow to lift workflow clarity to 5.

Dimension	Reasoning	Score
Conciseness	The ~1600-word body is operational rather than padded with known concepts, but routing rules are restated multiple times (the routing table, the 'Important compatibility behavior' bullets, and each backend section's 'explicit only' reminder), so it could be meaningfully tightened.	3 / 5
Actionability	Eight copy-paste bash invocations and a JSON context example, all using flags verified to exist in scripts/research_lookup.py, cover the academic, deep-research, chat, Perplexity, fast-lookup, and batch cases — fully executable.	5 / 5
Workflow Clarity	A clear 5-step numbered manuscript workflow with an explicit verification step (Extract) and feedback loops in Failure handling (shortfall -> inspect coverage.json -> refine), but the validate->fix->re-verify loop is not woven as crisply into the main sequence as the 5-anchor example.	4 / 5
Progressive Disclosure	Well-organized with clear headers and one-level references to real bundle scripts (research_lookup.py, manuscript_packet.py) with no nested references, but the single ~320-line file inlines the 10 quality rules and full flag coverage that could split into separate reference docs.	4 / 5
	Total	16 / 20 Passed

Description

88%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 strong: it pairs concrete actions with an explicit 'Use when...' trigger clause and names the specific backends and their roles. It is held back only by slightly incomplete synonym coverage and minor overlap with adjacent review/citation skills.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions and the specific backend each performs: 'Compile current scholarly evidence', 'gather literature, references, background evidence, competing findings, or a manuscript research packet', plus Parallel Search/Extract/Research/Chat/Perplexity each tied to a role — comprehensive concrete coverage matching the 5 anchor.	5 / 5
Completeness	Explicitly answers both what ('Compile current scholarly evidence for a scientific manuscript or research brief') and when ('Use when the user explicitly asks to gather literature, references, background evidence, competing findings, or a manuscript research packet') with concrete trigger phrases, matching the 5 anchor.	5 / 5
Trigger Term Quality	Good natural trigger coverage ('literature, references, background evidence, competing findings, manuscript research packet') that users would actually say, but misses common synonyms like 'papers', 'studies', 'citations', 'bibliography', or '.bib' that would lift it to comprehensive.	4 / 5
Distinctiveness Conflict Risk	A clear niche (manuscript evidence packets) with distinct backend-routed triggers, but minor overlap risk with closely related skills like literature-review and citation-management that the body itself cross-references.	4 / 5
	Total	18 / 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_field	'metadata' should map string keys to string values	Warning

	Total	15 / 16 Passed

Repository: K-Dense-AI/claude-scientific-writer
Path: .claude/skills/research-lookup/SKILL.md
Commit: 14ee3e3

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.