research

Conduct preliminary research on a topic and generate research outline. For academic research, benchmark research, technology selection, etc.

Quality

57%

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/research-en/research/SKILL.md

Quality

Content

65%Weight 40%Scale 1-3

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

The content is highly actionable thanks to an executable prompt template and a clear five-step workflow, and it largely avoids padding. Its main weaknesses are a redundant duplicated example, a missing validation checkpoint for the batch-agent step, and a monolithic structure that could offload the verbose template into a reference file.

Suggestions

Replace or condense the near-duplicate one-shot example so it only shows the variable substitutions rather than restating the entire template, or move the example into a reference file.

Add an explicit validation/confirmation checkpoint in Step 4 before writing outline.yaml and fields.yaml (e.g. review merged items and fields with the user, or sanity-check counts against batch_size).

Extract the full prompt template into a referenced file (e.g. references/web_search_prompt.md) and keep SKILL.md as a concise overview with a one-level-deep pointer.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence without explaining known concepts, but it duplicates the prompt template as a near-verbatim one-shot example (the example largely restates the template structure), adding tokens that could be trimmed.	2 / 3
Actionability	It provides a copy-paste-ready, fully specified web-search-agent prompt template with explicit variables, a hard constraint, structured output requirements, and a concrete one-shot example, matching the level-3 anchor for executable guidance.	3 / 3
Workflow Clarity	The five steps are clearly numbered with AskUserQuestion confirmation checkpoints, but the batch parallel-agent generation in Step 4 has no explicit validation/verification step before files are written, which per the guidelines caps batch-operation workflows at 2.	2 / 3
Progressive Disclosure	The single-file skill is organized into clear sections, but the bulky, largely-redundant prompt template plus one-shot example is inline content that could be split into a reference file, and navigation between the duplicated template and example is not signaled.	2 / 3
	Total	9 / 12 Passed

Description

50%Weight 40%Scale 1-3

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 clearly states the skill's purpose and gives useful example domains, but it stops short of an explicit trigger clause and a comprehensive set of natural trigger terms. It is functional yet reads more like a categorized capability statement than a sharply tuned activation description.

Suggestions

Add an explicit 'Use when...' clause, e.g. 'Use when the user wants to scope out a new topic, build a research outline, or compare options before a deep dive.'

Broaden trigger terms to natural phrasings users actually say, such as 'research outline', 'literature review', 'state of the art', or 'compare technologies'.

List a fuller set of concrete actions (e.g. 'survey items, propose a field framework, draft outline.yaml and fields.yaml') to raise specificity.

Dimension	Reasoning	Score
Specificity	The description names the research domain plus two concrete actions ('conduct preliminary research' and 'generate research outline'), but the actions are only lightly specified rather than a comprehensive list, matching the level-2 anchor and falling short of the multiple-specific-actions bar for 3.	2 / 3
Completeness	The 'what' is explicit (conduct research, generate outline), but the 'when' is only implied via example categories with no 'Use when...' trigger clause, which per the guidelines caps completeness at 2.	2 / 3
Trigger Term Quality	It lists several plausible user terms ('academic research', 'benchmark research', 'technology selection'), but these read as research categories rather than a broad coverage of natural phrasings users would say, and common variations (e.g., 'literature review', 'research outline') are missing.	2 / 3
Distinctiveness Conflict Risk	The 'preliminary research + outline' framing carves a niche distinct from a deep-research skill, yet the trigger context (academic/benchmark/technology selection) is broad enough to overlap with general web-search or deep-research skills, fitting the level-2 anchor.	2 / 3
	Total	8 / 12 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
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
relative_links	Relative link issues: 4 missing	Warning

	Total	14 / 16 Passed

Repository: Weizhena/Deep-Research-skills
Path: skills/research-en/research/SKILL.md
Commit: e5479f8

Reviewed: 11 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.